Nonconvulsive seizures (NCSzs) are electrographic seizures with little or no overt clinical manifestations (Fig. 54-1). The EEG criteria for definite NCSzs are outlined in Table 54-2. They are an increasingly commonly recognized entity in neuro-ICUs, where 8% to 34% of comatose patients may have NCSzs, depending on the clinical setting.¹^–¹¹ Major studies investigating NCSzs in various patient populations are summarized in Table 54-3. Nonconvulsive status epilepticus (NCSE) occurs when NCSzs are prolonged; a commonly used definition is continuous or nearly continuous electrographic seizures of at least 30 minutes’ duration.¹² Although some forms of NCSzs may occur in ambulatory patients (typically manifested as confusion and easily treated), our main focus in this chapter is on NCSzs in neuro-ICU patients, for whom the most common manifestation is a depressed level of consciousness.¹² Most patients with NCSzs have purely electrographic seizures,⁸ but other subtle signs that have been associated with NCSzs include face and limb myoclonus, nystagmus, eye deviation, pupillary abnormalities (including hippus), and autonomic instability.¹²^–¹⁵ However, none of these clinical findings are highly specific for NCSzs, and cEEG is usually necessary to confirm or refute the diagnosis of NCSzs. A recent study showed that findings from cEEG monitoring in 287 patients led to a change in AED prescribing in 52% of all studies, with initiation of AED therapy in 14%, modification of AED therapy in 33%, and discontinuation of AED therapy in 5% of all studies. The detection of electrographic seizures led to a change in AED therapy in 28% of all studies.^15a

FIGURE 54-1 Nonconvulsive seizures and status epilepticus, focal. A, This 49-year-old woman with a history of head trauma, seizures, and strokes was being evaluated for decreased mental status. During the recording, she had repetitive right-sided electrographic seizures beginning in the right frontal region (highlighted) and was in focal (also known as partial) nonconvulsive status epilepticus. B, Thirty seconds later, the discharge has become faster, thus demonstrating evolution in frequency and morphology and therefore unequivocally ictal. C, Four minutes later, the discharge continues and is becoming more widespread and slower, thus demonstrating evolution in location and further evolution in frequency and morphology. There were no associated movements.D, The discharge is now in the right temporal region only. The patient was given intravenous lorazepam. E, The discharge ceased after the administration of lorazepam. Beta activity (probably from the benzodiazepine) is present, but predominantly over the left hemisphere (box) (i.e., attenuated on the right—see ellipse—suggestive of cortical dysfunction on the right, as is commonly seen postictally); mild focal slowing (suggesting dysfunction, including subcortical white matter) can also be appreciated in the right parasagittal region (ellipse).

(From Hirsch LJ, Brenner RP. Atlas of EEG in Critical Care. London: Wiley-Blackwell, 2010.)

TABLE 54-2 Criteria * for Definite Nonconvulsive Seizure

Any pattern lasting at least 10 seconds and satisfying any one of the following three primary criteria:

Primary Criteria

Repetitive generalized or focal spikes, sharp waves, spike and wave complexes at ≥3/sec

Repetitive generalized or focal spikes, sharp waves, spike and wave or sharp and slow-wave complexes at <3/sec and the secondary criterion

Sequential rhythmic, periodic, or quasi-periodic waves at ≥1/sec and unequivocal evolution in frequency (gradually increasing or decreasing by at least 1/sec, e.g., 2-3/sec), morphology, or location (gradual spread into or out of a region involving at least two electrodes). Evolution in amplitude alone is not sufficient. Change in sharpness without other change in morphology is not enough to satisfy evolution in morphology

Secondary Criterion Significant improvement in the clinical state or appearance of previously absent normal electroencephalographic patterns (such as posterior-dominant “alpha” rhythm) temporally coupled to acute administration of a rapidly acting antiepileptic drug. Resolution of the “epileptiform” discharges leaving diffuse slowing without clinical improvement and without the appearance of previously absent normal electroencephalographic patterns would not satisfy the secondary criterion

* Satisfying these criteria is adequate for confirming nonconvulsive seizure activity. However, failing to meet these criteria does not rule out nonconvulsive seizure activity; clinical judgment and correlation are required in this situation.

Adapted from Chong DJ, Hirsch LJ: Which EEG patterns warrant treatment in the critically ill? Reviewing the evidence for treatment of periodic epileptiform discharges and related patterns. J Clin Neurophysiol. 2005;22:79, as modified from Young GB, Jordan KG, Doig GS. An assessment of nonconvulsive seizures in the intensive care unit using continuous EEG monitoring: an investigation of variables associated with mortality. Neurology. 1996;47:83.

TABLE 54-3 Studies Using Continuous Electroencephalographic Monitoring in Critically Ill Patients for Detection of Nonconvulsive Seizures

The causes of NCSzs and NCSE in neuro-ICU patients may vary by age and patient group but in general are similar to the causes of convulsive seizures in these patients, including structural lesions, infections, metabolic derangements, toxins, drug withdrawal, and epilepsy, all of which are common diagnoses in critically ill patients.¹⁶ However, nonconvulsive or subtle seizures are much more common than clinically overt seizures in critically ill patients (see Table 54-3). In this section we review the incidence of NCSzs/NCSE in neuro-ICU patients by diagnostic and demographic category.

Convulsive Status Epilepticus

In many patients with convulsive status epilepticus (SE), electrographic seizures can persist even when convulsive seizures have stopped.^17,¹⁸ In a prospective study, DeLorenzo and colleagues found that 48% of the patients monitored with cEEG for 24 hours after convulsive SE had stopped experienced NCSzs and 14% had NCSE.¹⁹ In most of these patients, coma was the only clinical manifestation. Patients with NCSE after convulsive SE had a twofold greater mortality than did patients whose seizures ended when convulsive activity stopped in this study, as well as in the Veterans Affairs Cooperative Study.¹⁸ Therefore, cEEG should be performed on any patient who does not quickly regain consciousness after a convulsive seizure to detect ongoing seizure activity. This includes patients who are sedated or paralyzed (or both) during the treatment of SE in whom the level of consciousness cannot be assessed adequately.

Subarachnoid Hemorrhage

Aneurysmal SAH has long been known to be associated with seizures. Studies have reported 4% to 9% rates of convulsive seizures after the initial bleeding,²⁰^–²³ often in the setting of a focal clot.^21,^23,²⁴ However, several more recent studies using cEEG suggest that these seizure rates underestimate the incidence of electrographic seizures after SAH, especially in comatose patients. In the Columbia series of 570 patients who underwent cEEG for altered mental status or suspicion of seizures, 19% of 108 SAH patients had seizures.⁸ Most of these seizures were NCSzs, and 70% of the patients with seizures had NCSE. In a study of 11 patients with SAH and NCSE, the following clinical factors were associated with NCSE: advanced age, female sex, need for ventriculostomy, poor-grade SAH (Hunt-Hess grade III, IV, or V), thick cisternal blood clots, and structural lesions (ICH and stroke).²⁵ Because both convulsive seizures and NCSzs are associated with a poor outcome on multivariate analysis in patients with a diagnosis of SAH,^20,²⁶ it may follow that seizures occurring after SAH are likely to worsen the brain injury (see the later section “Impact of Nonconvulsive Seizures in the Critically Ill”). However, the currently available data are insufficient to confirm a causal relationship.

Intracerebral Hemorrhage

ICH is associated with a 3% to 19% rate of in-hospital convulsive seizures.^10,^24,²⁷^–³⁰ In two recent studies using cEEG, 18% to 21% of patients with ICH were shown to have NCSzs.^7,¹⁰ cEEG may also predict outcome after ICH. Vespa and coworkers found that NCSzs were associated with an increased midline shift and a trend toward worse outcomes after controlling for hemorrhage size.⁷ In a study of patients with ICH by Claassen and associates, NCSzs were associated with expansion of hemorrhage volume and a trend toward worse outcomes as well.¹⁰ In addition, periodic epileptiform discharges (PEDs) were an independent predictor of poor outcome.

Ischemic Stroke

Population- and hospital-based studies have reported rates of acute clinical seizures after ischemic stroke ranging from 2% to 9%.^24,^27,^29,³¹^–³³ Again, more recent studies using cEEG have shown that this may be an underestimate. In the Columbia series, 11% of 56 patients with ischemic stroke undergoing cEEG had seizures; these seizures were purely nonconvulsive in 5 of these 6 patients.⁸ In a recent study using cEEG, 6% of 46 patients with ischemic stroke demonstrated nonconvulsive seizure activity.⁷ In the work of Jordan, who used cEEG in 57 consecutive patients admitted to the ICU with cerebral ischemia, 26% of the patients had EEG-defined NCSzs during the period of monitoring.³ Several studies have shown that acute clinical seizures are associated with increased mortality in patients with ischemic stroke.^24,^34,³⁵ The relationship between NCSzs and outcome after stroke is currently unknown. However, in a rodent model of acute stroke, NCSzs were associated with increased infarct volume and a threefold increase in mortality.³⁶

Traumatic Brain Injury

The incidence of convulsive seizures within the first week after TBI is 4% to 14%³⁷^–³⁹ and increases to 15% in patients with severe TBI.^37,³⁹ Data regarding the incidence of NCSzs after TBI are relatively scant, but rates from 18% to 28% have been reported.^1,^6,⁸ In 96 consecutive patients with moderate or severe TBI undergoing cEEG, Vespa found that 22% had seizures; more than half of the patients with seizures had NCSzs only, and many of these patients had therapeutic phenytoin serum levels.⁴⁰ In the Columbia series, 18% of the 51 patients with TBI monitored with cEEG had seizures, all of them had NCSzs, and 8% had NCSE.⁸ The exact relationship between seizures and outcome is unclear, but some studies have shown that early posttraumatic seizures are an independent risk factor for poor outcome in adults⁴¹ and children with severe TBI.⁴² Whether NCSzs have a similar impact has not been studied properly.

Postoperative Patients

Neurosurgical procedures, especially those involving supratentorial lesions, are associated with a 4% to 17% risk for postoperative clinical seizures.⁴³^–⁴⁶ Patients with a presurgical history of epilepsy have a risk for postoperative seizures that has been reported to be as high as 34%.⁴⁶ Little is known about the rate of NCSzs in these patients. In the Columbia cEEG series, 3 of 13 patients monitored after neurosurgery (excluding patients with SAH) had seizures, all NCSzs, and 1 had NCSE.⁸ Postoperative seizures can occur in any postoperative setting in which there is an acute neurological injury, a high risk for metabolic derangement, or neurotoxicity for any reason.

One particularly high-risk non-neurosurgical group is transplant patients. Seizures are common after pancreas, liver, lung, heart, kidney, and bone marrow transplantation ⁴⁷^–⁵⁷ and often occur in the immediate postoperative period. Patients undergoing cardiac surgery are at risk for the development of acute neurological complications, including stroke or hypoxia,⁵⁸ that may predispose them to seizures in the perioperative or postoperative period.⁵⁹ The incidence of NCSzs and NCSE in these patients has not been studied adequately.

Hypoxic-Ischemic Injury

In a series of comatose patients with NCSE, 42% of the patients had hypoxic/anoxic injury.⁵ Rates of seizures have been reported to be as high as 35% after cardiac arrest.^60,⁶¹ Twenty percent of the patients with hypoxic-ischemic injury monitored in the Columbia series had seizures, most of which were NCSzs.⁸ Aside from being a potential contributor to decreased mental status in these patients, the presence of seizures after cardiac arrest may have important prognostic implications (see the later section “Prognostication”).⁶² In addition, as hypothermia becomes more widely implemented for neuroprotection after cardiac arrest, cEEG may become an important tool for identifying NCSzs, especially during rewarming.⁶³

Toxic-Metabolic Encephalopathy

Critically ill patients are susceptible to many toxic, metabolic, and electrolytic imbalances that may cause both changes in mental status and seizures. Such conditions include but are not limited to hypoglycemia and hyperglycemia, hyponatremia, hypocalcemia, drug intoxication or withdrawal, uremia, liver dysfunction, hypertensive encephalopathy, and sepsis.¹⁶ In the Columbia series, 21% of the patients monitored with cEEG who had toxic-metabolic encephalopathy as their primary neurological diagnosis experienced NCSzs.⁸ In other series, 5% to 10% of patients with acute NCSzs had metabolic derangements as the probable cause of their seizures.^5,⁶⁴ In a recent study of 201 medical ICU patients without known brain injury who underwent cEEG, 22% had PEDs or seizures; sepsis and acute renal failure were significantly associated with electrographic seizures.⁶⁵

Impact of Nonconvulsive Seizures in the Critically Ill

As evidenced by the preceding discussion, NCSzs are clearly common in critically ill patients, and this prompts the question of whether NCSzs require rapid identification and treatment and whether this would have an impact on patient outcome. There is previous evidence that delay in diagnosis and the duration of NCSE are each independent predictors of a worse outcome, including higher mortality,^6,⁶⁴ although mortality in patients with NCSE may be most influenced by the underlying cause.⁶⁶ In addition, although NCSE may be associated with a poor prognosis in the critically ill elderly,⁶⁷ aggressive treatment of NCSzs and NCSE itself may be associated with worse outcomes in this population.⁶⁸ Definitive proof that NCSzs worsen outcomes is lacking, however; to date, there has not been a single prospective controlled trial to determine whether treating NCSzs or NCSE improves neurological outcomes. Therefore, much of the justification for identifying and treating NCSzs in the critically ill comes from human and animal data demonstrating that seizures can lead to neuronal injury.

There is a large body of evidence that prolonged seizures, including NCSzs, can lead to neuronal damage in several animal models. Meldrum and colleagues found that paralyzed and artificially ventilated baboons exhibited hippocampal cell loss after treatment with a convulsant.⁶⁹ Despite careful control of factors such as oxygenation, temperature, and metabolic status, cell death occurred after 60 minutes of continuous electrographic seizures. Electrical- and chemoconvulsant-induced SE in rodents leads to cell loss, free radical production, inflammation, gliosis, and synaptic reorganization.⁷⁰ Pathologic changes can be seen in the absence of overt convulsions and can have profound long-term effects such as impaired performance on cognitive tasks⁷¹ and the development of epilepsy.⁷² There is also some evidence from animal models that even single or multiple brief seizures may lead to cell death and cognitive impairment.^73,⁷⁴ SE in humans has likewise been associated with hippocampal cell loss in postmortem studies⁷⁵ and evidence of cell injury in hospitalized patients as demonstrated by elevated levels of serum neuron-specific enolase^76,⁷⁷ (NSE). Although the sequelae of NCSzs and NCSE are not as well understood, evidence suggests that they can lead to neuronal damage in humans. In a study of NSE levels after seizures, DeGiorgio and associates showed that NSE levels were especially high after NCSzs and seizures of partial onset and that elevations were seen even in absence of acute brain injury.⁷⁷

In addition to direct pathologic effects, seizures can place increased metabolic, excitotoxic, and oxidative stress on at-risk brain and lead to irreversible injury that may worsen the extent of the primary neurological insult. Microdialysis studies in patients with TBI have demonstrated that extracellular glutamate increases to excitotoxic levels after NCSzs.⁷⁸ Significant elevations in the ratio of lactate to pyruvate (suggesting neuronal stress and potential death), glycerol (suggesting membrane breakdown),⁷⁹ and intracranial pressure (ICP) are also seen.⁸⁰ As mentioned earlier, NCSzs in patients with ICH were associated with increased mass effect on serial imaging⁷ and expansion of hematoma size.¹⁰ NCSzs have been associated with increased infarct volumes after occlusion of the middle cerebral artery in rats,³⁶ treatment of which was shown to result in reduced infarct volumes.⁸¹ In addition, even brief seizures can lead to hemodynamic changes, such as increased cerebral blood flow (CBF), which may lead to transient and potentially injurious elevations in ICP, even in the absence of tonic-clonic activity.^82,⁸³ Hippocampal atrophy can be seen on long-term follow-up MRI ipsilateral to NCSz.⁸⁴

Periodic Patterns on Continuous Electroencephalography—What Do They Mean?

There are several periodic patterns commonly seen in critically ill patients in which the relationship to seizures is unknown.⁸⁵ Although certain periodic discharges may be more closely related to systemic metabolic abnormalities, such as triphasic waves in patients with hepatic encephalopathy, others reflect injured tissue at high risk for seizures, such as periodic lateralized epileptiform discharges (PLEDs, also known as lateralized periodic discharges⁸⁶). See Figure 54-2 for an example of PLEDs evolving into a focal seizure. Furthermore, periodic patterns are sometimes definitively ictal, such as when they are associated with time-locked contralateral clonic jerking. Periodic discharges are typically thought to be interictal or on an interictal-ictal continuum.⁸⁵ However, the evidence that PLEDs are occasionally ictal include the following: positron emission tomography in patients with frequent PLEDs demonstrates increased regional glucose metabolism similar to that seen with focal seizures,⁸⁷ single-photon emission computed tomography in patients with PLEDs demonstrates increased regional cerebral perfusion that normalizes when the PLEDs resolve,^88,⁸⁹ and there are reports of PLEDs causing reversible confusion that resolves with antiepileptic drugs. In one series, seven patients older than 60 years experienced recurrent confusional episodes associated with PLEDs on EEG with interdischarge intervals as long as 4 seconds.⁹⁰ The clinical deficits resolved with a slowing of the EEG discharges, whether spontaneously or prompted by intravenous benzodiazepines. Carbamazepine appeared to be effective in preventing recurrences.