Epidemiology of Seizures in Critically Ill Adults


Preexisting epilepsy

 Insufficient doses of AEDs

 Low levels of AEDs

 Sleep deprivation, stress, nonspecific illness

Cerebrovascular disorders

 Ischemic stroke

 SAH

 ICH

Traumatic brain injury

Hypoxic-anoxic injury

CNS infection

 Viral, bacterial, less common infectious etiologies

Sepsis

Brain tumors

Toxic-metabolic disorders and drugs

 Electrolyte disturbances

 Intoxication and withdrawal states (alcohol and illicit drugs)

 Medications

Inflammatory and immune-mediated conditions

 NORSE

 Paraneoplastic or non-paraneoplastic limbic encephalitis



In the literature, some studies seek to emphasize the importance of the underlying etiology of NCSE and NCS by dividing or categorizing patients into groups based on the associated diagnosis or precipitant. For example, acute symptomatic causes of NCSE are defined by acute neurologic or medical disorders causally related to the status, independent of a prior epilepsy diagnosis. Such conditions can include strokes, head trauma, hypoxia-anoxia, central nervous system (CNS) infections, or brain tumors [18]. Alternatively, patients with a preexisting diagnosis of idiopathic epilepsy who present in NCSE without having one of the acute symptomatic conditions would comprise an idiopathic group. Lastly, cryptogenic patients would be those without an idiopathic epilepsy diagnosis or an acute symptomatic cause of NCSE [18]. Thinking about NCSE in this manner helps to highlight the variations between these patient groups and the significance and influence that etiology can potentially have on factors such as treatment, prognosis, and outcome.

The expanding use of cEEG monitoring has aided in the evaluation of seizures and SE, especially NCSE, in critically ill adults. It has also lead to greater focus on identifying seizures in comatose patients who may be admitted to the ICU for reasons initially unrelated to epilepsy or seizures. Patients may undergo evaluation with cEEG in a variety of intensive care unit settings, including the medical ICU (MICU) or neurologic ICU (NICU). Often, cEEG monitoring is requested to evaluate impaired level of consciousness or altered mental status in such patients with multiple neurologic and medical comorbidities, which may put them at increased risk for seizures. Literature has suggested that the admission diagnoses and etiologies of seizures and SE may vary between patients admitted to a MICU as opposed to NICU [26]. For example, recent work has found a higher rate of toxic-metabolic, infectious, and substance-related etiologies in a MICU population and a higher rate of stroke in the NICU [26]. Overall, cEEG monitoring is a valuable tool in the detection and management of seizures and SE of varied etiologies in critically ill patients.


Preexisting Epilepsy


Adults with a preexisting diagnosis of epilepsy, either of focal or generalized onset, can present with an acute worsening of their seizures or with SE. This can occur due to one of the acute symptomatic etiologies discussed in detail in the following sections, such as an acute stroke, CNS infection, or a metabolic disturbance exacerbating the underlying epilepsy. However, in the case of preexisting epilepsy, alternate etiologies may include nonadherence to an AED regimen, low AED levels, or treatment with an inappropriate medication for a particular epilepsy type [24]. A potential example of the latter would be treatment of a primary generalized epilepsy syndrome, such as JME, with a medication such as carbamazepine that can actually exacerbate the underlying seizure disorder. In individuals with epilepsy, estimates in the literature suggest that around 15 % have developed SE at least once during the course of their illness [24]. Additionally, SE can also be the initial presentation of seizures in patients who will ultimately go on to develop epilepsy [6].

The particular presenting features of acute seizures and SE in epilepsy patients will vary based on the principal epilepsy diagnosis. For example, patients with an underlying primary generalized epilepsy may present with nonconvulsive, absence SE, or potentially GCSE, while those with a focal epilepsy syndrome can develop focal NCSE or GCSE [24]. MSE can also occur in the setting of preexisting epilepsy and in this case certainly represents a unique clinical scenario, as opposed to myoclonic status in patients with anoxic brain injury and cardiac arrest, which will be reviewed later. Examples of syndromes in which myoclonic status can develop include JME and secondary generalized epilepsies such as LGS.

In the setting of epilepsy, there are additional factors which can lower the threshold for seizures and possibly contribute to acute breakthrough seizures or SE in particular clinical scenarios. Such factors include significant sleep deprivation, alcohol or drug intoxication or withdrawal, stress, fatigue, nonspecific illness, or metabolic abnormalities [27]. These circumstances may precipitate acute seizures or SE in individuals with an underlying predisposition to seizures, and clinicians can often provide important education to patients on avoidance of at least some of these potential triggers.


Acute Cerebrovascular Injuries


Acute cerebrovascular injuries, including subarachnoid hemorrhage (SAH), ischemic stroke, and intracerebral hemorrhage (ICH), are common etiologies of seizures and SE in critically ill adults. In addition, remote hemorrhagic and ischemic strokes have also been recognized as frequent precipitants, especially in older adults [28]. A study that looked at causes of hospital-onset seizures found stroke to be one of the most common etiologies, accounting for 23 % of cases in those patients without a prior history of seizures [29]. Patients with acute cerebrovascular disorders may have convulsive seizures; nonconvulsive, electrographic seizures; or a combination of the two categories. In this patient population, seizures and SE can occur as part of the initial presentation of the brain injury or may be detected during the subsequent hospital course. The expansion of cEEG monitoring has aided in the detection of NCS and NCSE in patients with strokes and brain hemorrhages that may frequently be encephalopathic or comatose or may require prolonged sedation.

Recent work has demonstrated electrographic seizures in up to one third of patients with nontraumatic ICH undergoing cEEG monitoring, and over half of these seizures were not associated with any clear clinical findings [30]. Another group investigating rates of NCSE among patients referred for EEG testing found that out of 451 study subjects, ICH, including traumatic ICH, was the etiology in 18 % of cases [18]. In addition to electrographic seizures, periodic EEG patterns can also be seen in patients with ICH, including lateralized periodic discharges (LPDs), generalized periodic discharges (GPDs), and rhythmic delta activity (RDA). Periodic patterns have been noted to have an association with cortical, as opposed to deep, hemorrhages. Similarly, an expanding ICH volume (30 % expansion at 24 h follow-up CT scan) has been associated with electrographic seizures [30]. An ongoing challenge in this patient group, as is the case with many critically ill populations, lies in determining the impact of treating seizures and periodic patterns on overall prognosis and patient outcome.

Acute clinical seizures, as well as NCS and NCSE, are recognized complications of SAH and felt to be generally associated with poor outcomes. As noted, a high clinical index of suspicion is often needed to diagnose nonconvulsive ictal activity in these patients. Recent estimates of the frequency of electrographic seizures in SAH patients undergoing cEEG monitoring range from 7 to 19 % [31, 32]. Reports have additionally suggested that SAH patients at higher risk for NCSE include those with poor neurologic grade (Hunt and Hess grade 4 or 5 and Fisher grade 3 or 4) and older age [33]. Treatment of NCS and NCSE in these patients is aimed at prevention of secondary brain injury and associated complications. Quantitative EEG analysis has also shown promise as an additional tool for evaluation of delayed cerebral ischemia in SAH and is increasingly utilized at academic and other centers performing high volumes of cEEG monitoring. Details of quantitative EEG techniques will be reviewed in a later chapter.


Traumatic Brain Injury


Following TBI, seizures and SE can occur, and cEEG monitoring can assist in the detection of these events, especially in patients with altered mental status or coma. Important factors contributing to the occurrence of seizures include brain injury severity and the existence of hemorrhagic contusions [34]. In a series of twenty moderate-severe TBI patients with non-penetrating head injuries monitored prospectively with cEEG and cerebral microdialysis, ten were found to have electrographic seizures [34]. Of those ten patients, seven had SE. In this group, seizures were also found to be linked to persistent elevations in intracranial pressure as well as increases in the lactate/pyruvate ratio in cerebral microdialysis measurements [34], suggesting that seizures can have multifaceted effects on the brain in this population. In another large cohort of critically ill adults undergoing cEEG monitoring, 18 % of the included patients with a diagnosis of TBI had electrographic seizures [32]. In the TBI population, as with many of the other patient groups discussed here, the effect treatment of seizures has on overall patient outcomes, and neurologic function remains unknown. However, treatment of identified seizures does at least provide a potential mechanism to help in prevention of further brain injury and metabolic distress [34].


Central Nervous System Infections


Electrographic and clinical seizures can occur with CNS infections of various types in critically ill adults, including cases of viral and bacterial infections, as well as with less common infectious diseases. A well-recognized example is that of herpes simplex encephalitis (HSE), a potentially life-threatening condition in which seizures are common during the course of the illness. EEGs performed in patients with HSE often demonstrate lateralized findings, such as lateralized periodic discharges (LPDs) [35], which can be highly epileptogenic. cEEG monitoring can be useful in identifying electrographic seizures and periodic patterns in patients with various CNS infections [36].

In a study of critically ill adults admitted with CNS infections who were monitored with cEEG, 48 % had either electrographic seizures or PDs [36]. The largest group of patients had a viral etiology (68 %), followed by bacterial and then fungal or parasitic causes. As has been discussed with other etiologic groups, only a fraction of the patients with electrographic seizures (36 %) had an appreciable clinical correlate, again highlighting the concept that a high index of suspicion along with the use of cEEG monitoring is often necessary to detect these seizures [36].

There are additional rare infectious causes of SE that have been described, including uncommon bacterial and viral infections and prion diseases, about which less is known. Some examples of atypical bacterial infectious agents include Bartonella, Coxiella burnetii (Q fever), and neurosyphilis [24, 37]. Less common viral causes that have been reported can include West Nile encephalitis, JC virus, Parvovirus B19, and measles encephalitis, among others; and prion diseases such as Creutzfeldt-Jakob disease are also noted [24, 37]. Such atypical infectious causes of SE should be considered in the appropriate clinical context and in cases where standard initial work-up is unrevealing.


Toxic-Metabolic Disorders and Drug-Related Causes


Toxic-metabolic disorders are an additional cause of acute seizures and SE in critically ill adults. There are a number of metabolic disturbances that can be associated with seizures, including hyponatremia, hypoglycemia or hyperglycemia, hepatic encephalopathy, uremia and renal failure, and less commonly hypomagnesemia and hypocalcemia [24, 29, 38]. While such metabolic abnormalities are recognized as seizure precipitants, precise cut points for serum values below or above which seizures may occur have yet to be fully delineated [38]. In patients with seizures or SE related to metabolic disturbances, there may be an additional underlying cerebral lesion that may be remote, such as a prior stroke. In this case, the metabolic abnormality would further lower the threshold for acute seizures in patients who may already be predisposed. A study evaluating the etiologies, treatment, and outcomes of hospital-onset seizures found metabolic abnormalities to be one of the most commonly observed causes of seizures (20 % of cases). It was an especially prevalent etiology among those without a prior existing history of seizures [29]. An interesting link between metabolic disorders and SE is that of EPC and hyperosmolar nonketotic hyperglycemia. EPC has multiple, varied causes, and among them metabolic disorders, such as nonketotic hyperglycemia, are relatively common [24, 39]. As mentioned previously, there is most often a co-occurring structural brain lesion in addition to the metabolic abnormality in these cases of EPC [39].

Alcohol and drug intoxication and withdrawal states are also accepted as precipitants for acute symptomatic seizures in adults [38]. If alcohol withdrawal is to be implicated as the etiology, then the seizure must take place within 7–48 h of the patient’s last drink [38]. Withdrawal from certain medications, such as benzodiazepines, can also cause acute seizures. Drugs of abuse that can precipitate acute symptomatic seizures include cocaine and crack cocaine, certain stimulants and inhalants, and potentially hallucinogens such as phencyclidine (PCP) [38].

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 12, 2017 | Posted by in NEUROLOGY | Comments Off on Epidemiology of Seizures in Critically Ill Adults

Full access? Get Clinical Tree

Get Clinical Tree app for offline access