Non-Convulsive Status Epilepticus (NCSE): Definition, Recognition, Electroencephalographic Findings, and Diagnosis

Abstract

Non-convulsive status epilepticus (NCSE) is being increasingly recognized in many patient populations, especially in patients with altered mental status (AMS) admitted to the intensive care unit (ICU), where its frequency is the highest. Because of the paucity of signs and symptoms, the diagnosis of NCSE needs a prolonged electroencephalography (EEG) recording which may take more than 24 hours. Once the diagnosis is established, an urgent treatment should be initiated. This chapter will review the history and definition, the EEG, and clinical manifestations of NCSE and will also focus on the many factors including the newly described ictal-interictal continuum which all contribute to an increase in the frequency of NCSE.

1. Introduction

Convulsive status epilepticus (CSE) is a life-threatening condition with a mortality rate varying between 7.6 and 39% [1, 2, 3]. Data are variable for non-convulsive status epilepticus (NCSE) [4]. One study reported that NCSE is associated with high mortality (18%) and morbidity (39%), and mortality correlates with the underlying etiology of NCSE (such as acute medical conditions), the development of acute complications, and severe deterioration in the level of consciousness [5]. Another recent study examining patients with suspicion or diagnosis of NCSE (total of 54) admitted to the internal medicine ward reported a mortality rate of 37%. In this study, the mortality rate correlated with the presence of hypernatremia and atrial fibrillation [6].

In addition, NCSE clinical definitions and EEG data change over time. As early as 1996, Young GB stated that available classifications of status epilepticus (SE) were inadequate for such cases, and more studies were needed to know whether patients with NCSE benefit from early seizure detection and treatment [7]. In 2005, the American Society for Clinical Neurophysiology (ACNS) published proposals for EEG terminology for intensive care units (ICUs), which are currently widely used [8]. In 2013, “The Salzburg Colloquium Consensus Criteria for Non-Convulsive Status Epilepticus (mSCNC)” [9] proposed criteria for the diagnosis of NCSE [9]. Finally, in 2015, the International League Against Epilepsy (ILAE) Task Force on Classification of status epilepticus published the definition and classification of status epilepticus [10].

In this chapter, we will review the history and definition of NCSE and in which clinical situations it is encountered; the use of continuous EEG (cEEG) monitoring in the detection of NCSE and the controversial, periodic EEG patterns associated with NCSE will be included in this chapter as well as the newly described ictal-interictal continuum (IIC). This chapter will not discuss the epidemiology, etiology, or treatment of NCSE, SE in children, which are covered in other chapters of this book.

2. History of NCSE

Prolonged non-convulsive seizures (PNCS) have been recognized since the nineteenth century; PNCS were included within the rubric of SE during the Marseilles Colloquium in 1962 organized in France and headed by Pr Henri Gastaux and his colleagues [11]; the idea that there were status forms for every seizure was initiated during that colloquium. The recent International Classification of SE reflected the classification of seizure types reported during the 1962 colloquium [11]. However, this classification based on seizure type did not fully describe the many clinical manifestations that are especially important for NCSs/NCSE.

Early descriptions and studies of NCSE were reported by H Gastaut [11], Shorvon SD [12], Kaplan, PW [13], Kaplan, PW [14], Drislane, FW [15], and Walker M. et al. [16]. Since then, there has been a wealth of publications regarding NCSE from different parts of the world.

3. Definition

There are currently multiple clinical definitions of NCSE; all highlighting three important criteria.

Alterations in the level of consciousness (mild confusion to coma) and non-convulsive clinical manifestations, abnormal brain electrical activity on EEG [17, 18, 19], and response to treatment (usually a trial of benzodiazepines) are additional diagnostic criteria. However, this definition does not take into account:

1. NCSE cases refractory to treatment [20]

2. Lack of evidence-based EEG criteria for NCSE

3. Poorly defined pathophysiological significance of the controversial rhythmic, periodic EEG patterns frequently observed with NCSE in the ICU, such as lateralized periodic discharges (LPD), generalized periodic discharges (GPD), bilateral independent periodic discharges (BIPD), lateralized rhythmic delta activity (LRDA), stimulus-induced rhythmic, periodic, or ictal discharges (SIRPIDS).

4. NCSE: clinical manifestations and in which clinical situations?

Alteration in the level of consciousness in patients admitted to ICU should raise suspicion for NCS/NCSE. In one recent study, the authors followed 80 patients with impaired consciousness admitted to the ICU following ischemic stroke; among those, 31 patients (38.75%) suffered from NCSE, as shown by electroencephalography (EEG) monitoring. Low Glasgow Coma Scale (GCS) was reported in 18 patients (58.1%), lethargy in 6 (19.4%), stupor in 11 (35.5%), and coma in 14 (45.2%) [21].

In another study, Sutter et al. reported the following clinical signs and symptoms such as speech disturbances, discrete myoclonias of face or limbs, bizarre behavior, anxiety, agitation, delirium, and hallucinations as clinical manifestations of NCSE [22].

PW Kaplan also mentions that real and genuine symptoms and clinical manifestations of NCSE, such as lethargy and confusion, unresponsiveness and cataplexy, obtundation, mutism, laughing, and crying, were missed and misinterpreted by the treating physician and caregiver as other clinical diagnoses [19].

A high index of suspicion is necessary to diagnose NCSE; it is common in ICU patients with altered mental status (AMS), especially in patients showing fluctuations in the level of consciousness.

No specific clinical manifestation is typical for the diagnosis of NCSE if we accept the alteration in the level of consciousness, which is common in ICU patients; however, discrete motor twitches involving the face or extremities and lip-smacking have been frequently reported in NCSE. Factors potentially associated with NCSE in elderly patients with acute confusional state include female gender, abnormal ocular movements, and history of epilepsy (Figure 1) [23].

Figure 1.

Ref. [23].

5. Emergent EEG vs. continuous EEG (cEEG) monitoring in the diagnosis of NCSE?

Many authors questioned the use of emergent EEG or continuous EEG (cEEG) monitoring in the diagnosis of NCSE in the ICU. In an interesting prospective study, Narayanan et al. studied 210 consecutive patients with AMS admitted to ICU to calculate the frequency of NCSE in those patients. All patients were initially evaluated with a 60-min emergent EEG (EmEEG) and later by continuous EEG (cEEG) monitoring. The results of this study showed that, among the 210 patients with AMS, the diagnosis of NCSE was confirmed in 22 patients (10.5%), in 12 (55%) patients using EmEEG, and in an additional 10 (45%) patients using EEG monitoring for 24–48 hours; the authors concluded that EmEEG may be used to diagnose NCSE but half of patients will be missed using this method of detection [24].

As mentioned by Raould Sutter, the number of recorded convulsive and non-convulsive seizures increases remarkably during the first 2 days after the start of cEEG monitoring in ICU patients with AMS [22].

Using cEEG to monitor inpatients with altered levels of consciousness, Claassen et al. also reported that only 56% of seizures were observed within the first hour of EEG recording and 93% within the first 48 hours [25].

In a 3-year, prospective, hospital-based study, using continuous EEG monitoring and following patients in the intensive care unit (ICU) presenting with an altered mental status (AMS), we reported a prevalence rate of NCSE of 26% [26]. This prevalence (26%) was in the middle range. NCSE patients did not perform better than controls, outcome being worse with comatose NCSE.

De Lorenzo et al. recommended the use of cEEG monitoring following the treatment of CSE because as high as 48% may show persistent electrographic seizures and around 14% of patients present NCSE [27].

These studies highlight the need for cEEG monitoring to detect NCS/NCSE, as short-term or emergent EEG is ineffective in detecting all these critical events [28].

In summary, cEEG monitoring is used

• To detect NCSs/NCSE

• Manage and treat NCSs/NCSE with antiseizure medications (ASMs) in a timely manner

• Control costs in the ICU, prevent refractory NCSE (RNCSE) (seizures lasting more than 60 min and failure of two ASMs) associated with high rates of hospital complications, long stay in hospital, and increased mortality.

6. Factors contributing to an increase in the incidence of NCSE in epidemiological studies

6.1 American Clinical Neurophysiology Society’s Standardized Critical Care EEG Terminology: 2012 version

In 2012, a group of critical care EEG experts developed terminology for EEG patterns encountered in critically ill patients in ICU. These experts have re-examined terms with clinical connotations, often referring to metabolic encephalopathy, such as “triphasic waves,” which many clinicians believe have no relation to seizures. They also standardized the terminology of periodic and rhythmic EEG patterns, quantified interictal EEG discharges, and categorized background EEG activity. This EEG terminology was primarily intended to help EEG researchers specialized in ICUs but had greatly assisted in making the diagnosis of NCSE. The 2021 new version of the “American Clinical Neurophysiology Society’s Standardized Critical Care EEG Terminology” is very useful; it adds recent research findings and definitions of several new terms and clarifies a few definitions of old terms [29, 30].

6.2 Salzburg criteria of NCSE (2013)

Table 1 summarizes the Salzburg EEG criteria of NCSE [9]. These criteria have been implemented in clinical practice and found to have high diagnostic accuracy [31].

Patients without known epileptic encephalopathy

EDs > 2.5 Hz, or

EDs ≤ 2.5 Hz or rhythmic delta/theta activity (>0.5 Hz) AND one of the following:

EEG and clinical improvement after IV AEDa, or

Subtle clinical ictal phenomena during the EEG patterns mentioned above or

Typical spatiotemporal evolutionb

Patients with known epileptic encephalopathy

Increase in prominence or frequency of the features mentioned above, when compared to baseline with observable change in clinical state

Improvement of clinical and EEGa features with IV AEDs

Table 1.

Working clinical criteria for non-convulsive status epilepticus.

^a

If EEG improvement occurs without clinical improvement, or if fluctuation without definite evolution, this should be considered possible NCSE.

^b

Incrementing onset (increase in voltage and change in frequency) or evolution in pattern (change in frequency >1 Hz or change in location) or decrementing termination (voltage or frequency).

Modified from Kaplan [19]. EDs, epileptiform discharges (spikes, poly spikes, sharp-waves, and sharp-and-slow-wave complexes); IV AEDs: intravenous anti-epileptic drugs.

In summary, EEG epileptiform discharges (EDs) should have a frequency of more than 2,5 HZ to qualify for NCSs/NCSE (Figure 2a, b); alternatively, EDs or periodic discharges (PDs) of less than 2.5 HZ can qualify for NCSs/NCSE if they shows some evolution (evolving EDs/PDs; evolution in frequency, amplitude, morphology) and are accompanied by either subtle clinical ictal phenomena during the EEG pattern (myoclonia of face or limbs, abnormal ocular movements, nystagmus, etc.), or if they show some spatiotemporal evolution (extension of the EDs to other brain areas) or show an EEG and clinical response following an IV benzodiazepine trial (IV lorazepam and diazepam) (Figure 3a–c).

Figure 2.

(a) and (b) from our laboratory.

Figure 3.

(a), (b) and (c) from our laboratory.

6.3 ILAE definition and classification of NCSE (2015)

In 2015, the ILAE task force on the classification of status epilepticus (SE) revised the concept, definition, and classification of SE [10]. This group reached a consensus that treatment of convulsive status epilepticus (CSE) should be initiated after approximately 5 minutes. No guidelines have been defined for the treatment of NCSE due to insufficient data; however, the task force indicated that the time to initiate emergency treatment for absence status epilepticus should begin after 10–15 minutes of seizure activity [10].

This new definition and classification of SE provided a framework for the diagnosis, workup, and management of each patient suspected of having SE, particularly NCSE.

6.4 Other factors contributing to an increase in the incidence of NCSE

The emergence of new technologies allowing several days of EEG recording, the widespread use of continuous EEG monitoring, its availability in ICUs and emergency departments around the world, as well as the increase in the elderly population and awareness of NCSE among physicians have led to an increase in the incidence and recognition of NCSE over the past two decades.

7. Controversial, periodic EEG patterns (LPDs, GPDs, etc.)

These controversial EEG patterns are described in the chapter of this book, “EEG manifestations of status epilepticus.” We will briefly discuss the following EEG features and their involvement in NCSs/NCSE: lateralized periodic discharges (LPDs) and generalized periodic discharges (GPDs).

Focal motor seizures are the most common LPD clinical manifestations [32]; the seizures can occur prior to or at the same time as LPDs [33]. In patients with LPDs, the risk of developing subsequent seizures varies from 10–56% [34]; Claassen et al. have shown that LPDs are highly associated with non-convulsive seizures (NCSs) (40%) [25] (Figure 4a, b); the frequency of LPDs correlates with seizure risk [35], with 40% of seizures for LPDs of less than 1 Hz and 66% of seizures with LPDs of 2 Hz or greater. The association of LPDs with acute traumatic brain injury is strongly associated with post-traumatic epilepsy, as is the association of LPDs with ischemic stroke, representing a high risk for vascular epilepsy [36].

Figure 4.

(a) and (b) from our laboratory.

At the start of cEEG monitoring in patients admitted to ICU and presenting with AMS, the majority of those with generalized periodic discharges (GPDs) are in a coma (>55%) [37]. GPDs are strongly associated with seizures [38]; NCSs are reported in 27% and NCSE in 22% of patients with GPDs [22, 38] (Figure 5a–c). GPDs may manifest as an ictal rhythm: NCSE may manifest as GPDs with triphasic morphology; in such cases, differentiation between interictal patterns, patterns associated with seizures, and patterns representing non-convulsive status epilepticus can be a real challenge [39]; these discharges may respond to benzodiazepines (Figure 6a, b).

Figure 5.

(a), (b) and (c) from our laboratory.

Figure 6.

(a) and (b) from our laboratory.

8. Ictal-interictal continuum (IIC)

There is no consensus regarding the definition of the ictal-interictal continuum (IIC); this is “a physiopathological and dynamic state where acute metabolic disturbances and instable neurobiological processes may or may not be involved in the generation of convulsive or non-convulsive seizures or status” (Figure 7) [40].

Figure 7.

Ictal-interictal continuum (IIC).

IIC includes most of the controversial EEG patterns (LPD, GPD, BIPD, LRDA, and GRDA), as well as other rhythmic, periodic patterns. These patterns are considered interictal when they show a frequency of less than 1 Hz or rhythmic/periodic discharges with a frequency of 1–2.5 Hz without spatiotemporal evolution and without clinical correlations and NCSs and NCSE when they exhibit a frequency of more than 2.5 Hz and manifest a spatiotemporal evolution and a clinical correlation.

There is an increased risk of seizures associated with IIC rhythmic patterns [41]; their morphology, duration, and frequency are predictive of the risk of seizures with a higher risk with IIC models with “plus” characteristics (modifiers) such as “LPDs + R” (R = rhythmic) or “LPD + F” (F = fast activity).

There are no data regarding the treatment of patients showing such EEG features; benzodiazepine (IV lorazepam) may be initiated, and EEG and clinical improvement can be monitored.

9. Conclusion

Non-convulsive status epilepticus (NCSE) is an emerging condition and requires, as for convulsive status epilepticus, an urgent diagnosis and treatment to prevent the occurrence of refractory NCSE. Its clinical manifestation consists mainly of an alteration in patient’s level of consciousness and an EEG showing characteristics features compatible with NCSE; a response to a benzodiazepine trial confirms the diagnosis; however, due to its multiple clinical presentations, a high index of suspicion is necessary to make the diagnosis of NCSE. Continuous EEG (cEEG) monitoring is necessary to diagnose and manage NCSE.