Seizures in the critically ill

Introduction

The routine use of continuous electroencephalographic (cEEG) recordings in intensive care demonstrates that clinical seizures in critically ill patients, both with neurologic and nonneurologic conditions, are the tip of the iceberg regarding seizure frequency in this vulnerable population. Nonconvulsive seizures are detected in 10%–30% of critically ill patients with altered mental status and in up to 60% of critically ill patients who have been witnessed to have prolonged clinical seizures earlier during their hospital stay. The presence of seizures in this population is associated with worse functional outcome at discharge, multiple markers of disease severity, and long-term neurologic injury. The routine use of cEEG monitoring and quantitative EEG (qEEG) analysis leads to the diagnosis of electrographic seizures and timely modification of treatment, which may affect early and long-term disability and mortality. The use of cEEG has become standard care in the intensive care unit (ICU). Its use is strongly recommended in cases of recent status epilepticus, unexplained coma, altered mental status, and behavioral changes and also in most patients with intracranial hemorrhage, as these cases are more likely to be associated with subclinical seizures.

Among patients without primarily neurologic conditions, the presence of a seizure may be the first indication of a central nervous system (CNS) complication; delay in recognition and treatment of seizures is associated with an increased risk of mortality. Thus rapid diagnosis, treatment, and follow-up of epileptic activity and seizures are crucial. In addition, because epilepsy affects around 2% of the population, patients with a predisposition to seizures may enter the ICU for treatment of other systemic problems. These patients are at high risk for developing seizures and status epilepticus in the context of acute or severe illness. Patients developing status epilepticus often require critical care management as treatment escalates to include multiple antiseizure medications and sedatives that require airway protection, hemodynamic support, and close monitoring.

Status epilepticus refers to prolonged seizure activity beyond the usual duration of a self-limited, isolated seizure; it may be the primary indication for admission to the ICU, or it may occur in ICU patients. Seizures that persist longer than 5 minutes or recurrent seizures without recovery to the patient’s neurologic baseline between seizures should be considered and treated as status epilepticus. Status epilepticus can also be subclinical, with little or no overt signs of seizures. Of patients monitored with EEG as part of a coma evaluation, 8%–10% were reported to be in nonconvulsive status epilepticus (NCSE). Recognizing the risk factors, clinical presentation, diagnosis, and management of clinical and subclinical seizures in the ICU is a frequent task for the intensivist. Recent technologic advances enormously facilitate the diagnosis and follow-up of these patients in real time. Communication between the ICU team and the neurology/neurophysiology team is crucial to offer the patient timely therapy that significantly affects outcome.

Epidemiology

Limited data are available on the epidemiology of seizures in the ICU. A 10-year retrospective study of all ICU patients with seizures revealed that 7 patients had seizures per 1000 ICU admissions. A 2-year prospective study of medical ICU patients identified 35 with seizures per 1000 admissions. In a retrospective analysis of 570 patients undergoing cEEG monitoring, seizures were detected in 110 patients (19%). Of those patients with seizures, the majority were in ICUs at the time of the EEG ( n = 101; 89%). In this series, 92% of the recorded seizures were nonconvulsive, requiring EEG for the diagnosis. Up to 34% of hospital inpatients experiencing a seizure die during their hospitalization. In a study of medical ICU patients, having even one seizure while in the ICU for a nonneurologic reason doubled in-hospital mortality. Incidence estimates for status epilepticus in the United States and Europe vary from 10.3 to 41/100,000. ^, The incidence has a bimodal distribution, peaking in young children and in the elderly. ^,

Mortality estimates vary widely, but data suggest that prolonged seizure duration is a negative prognostic factor. For patients whose generalized convulsive status epilepticus (GCSE) resolves within 30–59 minutes, the reported mortality is 2.7%, whereas patients seizing for more than 60 minutes have a mortality of 32%. If GCSE develops de novo in hospitalized patients, the mortality is 61%. For patients in NCSE, the mortality has been reported as high as 57% and correlates with the underlying cause, severe impairment of mental status, and development of complications, especially respiratory failure and infection. Determining the contribution of status epilepticus to mortality is challenging because of the large impact of the underlying etiology on outcomes.

Table 48.1 summarizes the most common causes of status epilepticus in adults in the community. Almost 50% of the cases were attributed to cerebral vascular disease. Antiseizure medication nonadherence is the main cause of status epilepticus in patients with a previous history of epilepsy; CNS infection, stroke, and metabolic disturbances predominate in patients without previous seizures.

Limited data are available concerning the functional abilities of GCSE survivors, and no data reliably permit distinction between the effects of status epilepticus and the effects of its causes. GCSE has the worst prognosis for neurologic recovery. Causes associated with increased mortality include anoxia, intracranial hemorrhages, tumors, infections, and trauma. In addition, intellectual ability declines as a consequence of status epilepticus. Survivors of status epilepticus often have memory and behavioral disorders out of proportion to the structural damage produced by the cause of their seizures. Case reports of severe memory deficits after prolonged focal status epilepticus with impairment of consciousness have been published. Conversely, one prospective study of 180 children with febrile status epilepticus reported no deaths or new cognitive or motor disability. Experimental animal and human epidemiologic studies suggest that status epilepticus may be a risk factor in the development of future seizures. Whether treatment of prolonged seizures reduces the risk of subsequently developing epilepsy remains uncertain.

Classification

Seizures are the clinical manifestation of the underlying predisposition for abnormal excessive and synchronous brain activity that defines epilepsy. Seizures can affect all measurable brain functions and follow a predictable clinical presentation largely related to the site of origin of the epileptic activity and the pattern of propagation.

The most recent revision of the terminology and classification of seizures and epilepsy from the International League Against Epilepsy (ILAE) was published in 2017 ( Table 48.2 ). Seizures are initially divided into focal- versus generalized-onset or unknown-onset seizures. Focal seizures are then further subdivided into focal-aware or focal-impaired awareness seizures. Additional classifiers for motor onset and nonmotor onset may be added. The motor classifiers include automatisms, atonic, clonic, epileptic spasms, hyperkinetic, myoclonic, and tonic, whereas nonmotor classifiers are autonomic, behavior arrest, cognitive, emotional, and sensory. Similarly, generalized-onset seizures are classified based on motor (e.g., tonic-clonic, myoclonic, atonic, and others) or nonmotor (i.e., typical absence, atypical absence, myoclonic) features. ^{, ,}

The ILAE terminology classifies epilepsy types based on varying levels of diagnosis: seizure type, epilepsy type, and epilepsy syndrome. Based on the three seizure types (focal, generalized, or unknown), the epilepsy type is determined as either focal, generalized, combined generalized and focal, or unknown. Potential etiologies should be considered, which include structural, genetic, infectious, metabolic, immune, or unknown etiologies. Epilepsy syndromes such as juvenile myoclonic epilepsy or childhood absence epilepsy refer to a cluster of features (seizure type, EEG, and imaging findings) that typically occur together. Determining the epilepsy type may aid in the selection of treatment options and prognosis. Careful description of the phenomenology of seizures, etiology, and localization continues to be an important part of the patient’s care. The identification of clinical features of seizures can be challenging in the ICU, except for observation of motor manifestations. ICU patients often have altered awareness caused by medications, hypotension, sepsis, or intracranial lesions; thus the presentation of focal seizures may be difficult to establish by observation alone, and EEG monitoring is often the only way to determine if a patient is seizing.

An ICU patient who has continuous, regular repetitive contractions in a stable muscle group in the right arm while awake and responsive can be classified as having focal right arm clonic status epilepticus without impairment of consciousness or awareness. Epilepsia partialis continua, a special form of focal status epilepticus in which repetitive movements of epileptic origin affect a small area of the body, can sometimes continue relentlessly for months or years and is notoriously refractory to treatment. The ILAE continues to work toward revising and updating the current classification system. For the most recent information, refer to www.ilae.org .