Status epilepticus with prominent motor symptoms clusters into distinct electroclinical phenotypes

Remote seizures and drug-resistant epilepsy after a first status epilepticus in adults

BACKGROUND AND PURPOSE

Long-term consequences after status epilepticus (SE) represent an unsettled issue. We investigated the incidence of remote unprovoked seizures (RS) and drug-resistant epilepsy (DRE) in a cohort of first-ever SE survivors.

METHODS

A retrospective, observational, and monocentric study was conducted on adult patients (age ≥ 14 years) with first SE who were consecutively admitted to the Modena Academic Hospital, Italy (September 2013-March 2022). Kaplan-Meier survival analyses were used to calculate the probability of seizure freedom following the index event, whereas Cox proportional hazard regression models were used to identify outcome predictors.

RESULTS

A total of 279 patients were included, 57 of whom (20.4%) developed RS (mean follow-up = 32.4 months). Cumulative probability of seizure freedom was 85%, 78%, and 68% respectively at 12 months, 2 years, and 5 years. In 45 of 57 patients (81%), the first relapse occurred within 2 years after SE. The risk of RS was higher in the case of structural brain damage (hazard ratio [HR] = 2.1, 95% confidence interval [CI] = 1.06-4.01), progressive symptomatic etiology (HR = 2.7, 95% CI = 1.44-5.16), and occurrence of nonconvulsive evolution in the semiological sequence of SE (HR = 2.9, 95% CI = 1.37-6.37). Eighteen of 57 patients (32%) developed DRE; the risk was higher in the case of super-refractory (p = 0.006) and non-convulsive SE evolution (p = 0.008).

CONCLUSIONS

The overall risk of RS was moderate, temporally confined within 2 years after the index event, and driven by specific etiologies and SE semiology. Treatment super-refractoriness and non-convulsive SE evolution were associated with DRE development.

The risk of unprovoked seizure occurrence after status epilepticus in adults

OBJECTIVE

Status epilepticus (SE) may lead to long-term consequences. This study evaluated the risk and predictors of seizure occurrence after SE, with a focus on SE due to acute symptomatic etiologies.

METHODS

Prospectively collected data about adults surviving a first non-hypoxic SE were reviewed. The outcome was the occurrence of unprovoked seizures during the follow-up. Kaplan-Meier survival curve analysis and log-rank test were used to analyze the time to seizure occurrence and determine the statistical significance between etiological groups. Three subcategories within acute etiology were considered according to the presence of the following: (1) structural lesion (acute-primary); (2) brain involvement during systemic disorders (acute-secondary); and (3) drug or alcohol intoxication/withdrawal (acute-toxic). Cox proportional hazards model was adopted to estimate hazard ratios (HRs) with the 95% confidence intervals (CIs).

RESULTS

Two hundreds fifty-seven individuals were included. Fifty-four subjects (21.0%) developed seizures after a median of 9.9 (interquartile range 4.3-21.7) months after SE. The estimated 1-, 2-, and 5-year rates of seizure occurrence according to acute SE etiologies were 19.4%, 23.4%, and 30.1%, respectively, for acute-primary central nervous system (CNS) pathology; 2.2%, 2.2%, and 8.7%, respectively, for acute-secondary CNS pathology; and 0%, 9.1%, and 9.1%, respectively, for acute-toxic causes. Five-year rates of seizure occurrence for non-acute SE causes were 33.9% for remote, 65.7% for progressive, and 25.9% for unknown etiologies. In multivariate Cox regression model, progressive etiology (adjusted HR [adjHR] 2.27, 95% CI 1.12-4.58), SE with prominent motor phenomena evolving in non-convulsive SE (adjHR 3.17, 95% CI 1.38-7.25), and non-convulsive SE (adjHR 2.38, 95% CI 1.16-4.90) were independently associated with higher hazards of unprovoked seizures. Older people (adjHR .98, 95% CI .96-.99) and people with SE due to acute-secondary CNS pathology (adjHR .18, 95% CI .04-.82) were at decreased risk of seizure occurrence.

SIGNIFICANCE

SE carries a risk of subsequent seizures. Both the underlying cause and epileptogenic effects of SE are likely to contribute.

How much refractory is 'refractory status epilepticus'? A retrospective study of treatment strategies and clinical outcomes

BACKGROUND AND PURPOSE

This study aimed to evaluate whether differences in clinical outcomes exist according to treatments received and seizure activity resolution in patients with refractory status epilepticus (RSE).

METHODS

Consecutive episodes of non-hypoxic status epilepticus (SE) in patients ≥ 14 years old were included. Episodes of RSE were stratified in: (i) SE persistent despite treatment with first-line therapy with benzodiazepines and one second-line treatment with antiseizure medications (ASMs), but responsive to successive treatments with ASMs (RSE-rASMs); (ii) SE persistent despite treatment with first-line therapy with benzodiazepines and successive treatment with one or more second-line ASMs, but responsive to anesthetic drugs [RSE-rGA (general anesthesia)]. Study endpoints were mortality during hospitalization and worsening of modified Rankin Scale (mRS) at discharge.

RESULTS

Status epilepticus was responsive in 298 (54.1%), RSE-rASMs in 152 (27.6%), RSE-rGA in 46 (8.3%), and super-refractory (SRSE) in 55 (10.0%) out of 551 included cases. Death during hospitalization occurred in 98 (17.8%) and worsening of mRS at discharge in 287 (52.1%) cases. Multivariable analyses revealed increased odds of in-hospital mortality with RSE-rGA (odds ratio [OR] 3.05, 95% confidence interval [CI] 1.27-7.35) and SRSE (OR 3.83, 95%. CI 1.73-8.47), and increased odds of worsening of mRS with RSE-rASMs (OR 2.06, 95% CI 1.28-3.31), RSE-rGA (OR 4.44, 95% CI 1.97-10.00), and SRSE (OR 13.81, 95% CI 5.34-35.67).

CONCLUSIONS

In RSE, varying degrees of refractoriness may be defined and suit better the continuum spectrum of disease severity and the heterogeneity of SE burden and prognosis.

Acute symptomatic status epilepticus: Splitting or lumping? A proposal of classification based on real-world data

This study aimed to group acute symptomatic etiologies of consecutive episodes of status epilepticus (SE) into different subcategories and explore their associations with clinical outcome. Etiologies were first categorized as "acute," "remote," "progressive," "SE in defined electroclinical syndromes," and "unknown." Four subcategories of acute etiologies were then defined: (1) withdrawal, low levels, or inappropriate prescription of antiseizure medications, or sleep deprivation in patients with pre-existing epilepsy; (2) acute insults to central nervous system (CNS; "acute-primary CNS"); (3) CNS pathology secondary to metabolic disturbances, systemic infection, or fever ("acute-secondary CNS"); and (4) drug/alcohol intoxication or withdrawal. Poor outcome at discharge, defined as worsening of clinical conditions (modified Rankin Scale [mRS] at discharge higher than mRS at baseline), was reported in 55.6% of cases. The etiological categories of acute-primary CNS (odds ratio [OR] = 3.61, 95% confidence interval [CI] = 2.11-6.18), acute-secondary CNS (OR = 1.80, 95% CI = 1.11-2.91), and progressive SE (OR = 2.65, 95% CI = 1.57-4.47), age (OR = 1.05, 95% CI = 1.04-1.06), nonconvulsive semiology with coma (OR = 3.06, 95% CI = 1.52-6.17), and refractoriness (OR = 4.31, 95% CI = 2.39-7.77) and superrefractoriness to treatment (OR = 8.24, 95% CI = 3.51-19.36) increased the odds of poor outcome. Heterogeneity exists within the spectrum of acute symptomatic causes of SE, and distinct etiological subcategories may inform about the clinical outcome.

Comparison of the status epilepticus severity score and the epidemiology-based mortality score in predicting 30-day mortality and status epilepticus cessation

OBJECTIVE

To evaluate the role of the Status Epilepticus Severity Score (STESS) and the Epidemiology-based Mortality score (EMSE) in predicting 30-day mortality and SE (Status epilepticus) cessation, and their prognostic performance in subgroups of patients with specific characteristics.

METHODS

We reviewed consecutive episodes of SE occurring in patients aged ≥14 years at Baggiovara Civil Hospital (Modena, Italy) from 2013 to 2021. We evaluated the predictive accuracy of EMSE and STESS for 30-day mortality and SE cessation through stepwise regression binary logistic models adjusted for possible univariate clinical confounders.

RESULTS

Seven hundred and eleven patients were enrolled. The mean value of STESS was 3.2 (SD 1.7) and of EMSE was 80.1 (SD 52.6). Within 30 days of the onset of SE, 28.4% of patients (202/711) died. EMSE had higher discriminatory ability for 30-day mortality compared with STESS (AUROC: 0.799; 95% CI: 0.765-0.832 versus 0.727; 95% CI: 0.686-0.766, respectively; p = 0.014). SE cessation within 1 h for convulsive SE and within 12 h for nonconvulsive SE was achieved in 35.3% (251/711) of patients. No significant difference was found between EMSE and STESS in discriminatory ability for SE cessation (AUROC: 0.516; 95% CI: 0.488-0.561 and 0.518; 95% CI: 0.473-0.563, respectively; p = 0.929). EMSE was superior to STESS in predicting 30-day mortality in patients with specific characteristics. No difference between the two scores was found in predicting SE cessation in subgroups of patients with specific characteristics.

CONCLUSIONS

EMSE seems superior to STESS in predicting 30-day mortality, particularly in specific patient categories. Conversely, there is no difference in the ability of these scores in predicting SE cessation, which is overall rather low.

Neurofilament light (NfL) as biomarker in serum and CSF in status epilepticus

OBJECTIVE

We explored the potential of neurofilament light chain (NfL) in serum and cerebrospinal fluid as a biomarker for neurodestruction in status epilepticus.

METHODS

In a retrospective analysis, we measured NfL in serum and cerebrospinal fluid samples of patients with status epilepticus using a highly sensitive single-molecule array technique (Simoa). Status epilepticus was diagnosed according to ILAE criteria. Additionally, we employed an alternative classification with more emphasis on the course of status epilepticus. We used data from three large control groups to compare NfL in status epilepticus versus neurologically healthy controls.

RESULTS

We included 28 patients (mean age: 69.4 years, SD: 15 years) with a median status duration of 44 h (IQR: 80 h). Twenty-one patients (75%) suffered from convulsive status epilepticus and seven (25%) from non-convulsive status epilepticus. Six patients died (21%). Cerebrospinal fluid and serum NfL concentrations showed a high correlation (r = 0.73, p < 0.001, Pearson). The main determinant of NfL concentration was the status duration. NfL concentrations did not differ between convulsive status epilepticus and convulsive status epilepticus classified according to the ILAE or to the alternative classification without and with adjusting for status duration and time between status onset and sampling. We found no association of NfL concentration with death, treatment refractoriness, or prognostic scores.

CONCLUSION

The results suggest that neurodestruction in status epilepticus measured by NfL is mainly determined by status duration, not status type nor therapy refractoriness. Therefore, our results suggest that regarding neurodestruction convulsive and non-convulsive status epilepticus are both neurological emergencies of comparable urgency.

Clinical scores and clusters for prediction of outcomes in status epilepticus

Status epilepticus (SE) is a life-threatening condition and may have long-term negative sequelae. Short- and long-term outcomes encompass mortality, deterioration of functional status compared to baseline, refractoriness to treatment, recurrence of SE, and development of epilepsy, cognitive impairment, and behavioral disturbances. So far, the greatest amount of evidence is available for the prediction of short-term mortality. Conversely, the knowledge regarding long-term consequences among SE survivors is still scarce and several issues have not yet been resolved. The heterogeneity of SE renders the prognostication of outcomes challenging. Although aetiology is the main determinant of the outcome, different prognostic predictors have been identified. In this regard, data on group effects need to be integrated into prognostic scores to allow individual risk stratification. Importantly, many of the present scores are not designed to enable repetition to follow patient evolution. A new paradigm for the assessment of SE outcomes should consider variables that become available and/or can be retested during the course of SE. Neuroimaging findings, serum biomarkers, treatment characteristics, complications during SE, peri-ictal and postictal characteristics after SE cessation look as promising determinants of outcome and are suitable for inclusion in future models to enhance the quality and increase the reliability of prediction. This paper was presented at the 8th London-Innsbruck Colloquium on Status Epilepticus and Acute Seizures held in September 2022.

The etiology and mortality of new-onset refractory status epilepticus (NORSE) in adults: A systematic review and meta-analysis

New-onset refractory status epilepticus (NORSE) is a devastating neurological presentation. There is a paucity of large studies on NORSE as it is a relatively new clinical syndrome. The aim of this review was to summarize the etiologies and establish a mortality rate for NORSE. Two independent authors systematically searched the following electronic databases from January 1, 2005 April 20, 2021: PubMed, Embase, OVID, Scopus, Web of Science, "Clinicaltrials.gov," and the International Standard Randomised Controlled Trial Number (ISRCTN) registry. We included all primary research studies of NORSE in adults and excluded commentaries, reviews, pre-clinical studies, and pediatric populations. Etiology was extracted from all studies meeting eligibility criteria, whereas data relating to treatments, hospital stay, functional outcomes, and mortality were extracted from studies with sample size ≥5. We conducted a random-effects meta-analysis of mortality rate with meta-regression testing for significant covariates. Of 1482 studies, 109 case reports and case series met our criteria, comprising 395 cases of NORSE. The most common etiology was cryptogenic in 197 cases (49.9%), followed by autoimmune in 143 cases (36.2%). The pooled mortality rate was 22% (95% confidence interval 17%-27%; N studies  = 15), with low heterogeneity ( I 2  = 0%). Meta-regression revealed that year of study, treatment with ketogenic diet or immunotherapy, percentage of cryptogenic cases, and length of intensive care unit stay were not significant covariates for mortality. Common treatments included antiseizure medications (median 5), general anesthesia, and immunotherapy such as corticosteroids, intravenous immunoglobulin, and plasma exchange. Mean length of intensive care admission was 33.4 days, with 52% of cases diagnosed with epilepsy on discharge. Neurocognitive impairment was a common sequela of NORSE. NORSE is associated with a high mortality. Half of cases remain cryptogenic, which presents a diagnostic challenge. Future focus should be on elucidating the underlying neurobiology and determining the most effective therapeutic interventions.

Retrospective External Validation of the Status Epilepticus Severity Score (STESS) to Predict In-hospital Mortality in Adults with Nonhypoxic Status Epilepticus: A Machine Learning Analysis

BACKGROUND

The objective of this study was to validate the value of the Status Epilepticus Severity Score (STESS) in the prediction of the risk of in-hospital mortality in patients with nonhypoxic status epilepticus (SE) using a machine learning analysis.

METHODS

We included consecutive patients with nonhypoxic SE (aged ≥ 16 years) admitted from 2013 to 2021 at the Modena Academic Hospital. A decision tree analysis was performed using in-hospital mortality as a dependent variable and the STESS predictors as input variables. We evaluated the accuracy of STESS in predicting in-hospital mortality using the area under the receiver operating characteristic curve (AUROC) with 95% confidence interval (CI).

RESULTS

Among 629 patients with SE, the in-hospital mortality rate was 23.4% (147 of 629). The median STESS in the entire cohort was 2.9 (SD 1.6); it was lower in surviving compared with deceased patients (2.7, SD 1.5 versus 3.9, SD 1.6; p < 0.001). Of deceased patients, 82.3% (121 of 147) had scores of 3-6, whereas 17.7% (26 of 147) had scores of 0-2 (p < 0.001). STESS was accurate in predicting mortality, with an AUROC of 0.688 (95% CI 0.641-0.734) only slightly reduced after bootstrap resampling. The most significant predictor was the seizure type, followed by age and level of consciousness at SE onset. Nonconvulsive SE in coma and age ≥ 65 years predicted a higher risk of mortality, whereas generalized convulsive SE and age < 65 years were associated with a lower risk of death. The decision tree analysis using STESS variables correctly classified 90% of survivors and 34% of nonsurvivors after the SE, with an overall risk of error of 23.1%.

CONCLUSIONS

This validation study using a machine learning system showed that STESS is a valuable prognostic tool. The score appears particularly accurate and effective in identifying patients who are alive at discharge (high negative predictive value), whereas it has a lower predictive value for in-hospital mortality.

Machine learning validation through decision tree analysis of the Epidemiology-Based Mortality Score in Status Epilepticus

OBJECTIVE

This study was undertaken to validate the accuracy of the Epidemiology-Based Mortality Score in Status Epilepticus (EMSE) in predicting the risk of death at 30 days in a large cohort of patients with status epilepticus (SE) using a machine learning system.

METHODS

We included consecutive patients with SE admitted from 2013 to 2021 at Modena Academic Hospital. A decision tree analysis was performed using the 30-day mortality as a dependent variable and the EMSE predictors as input variables. We evaluated the accuracy of EMSE in predicting 30-day mortality using the area under the receiver operating characteristic curve (AUC ROC), with 95% confidence interval (CI). We performed a subgroup analysis on nonhypoxic SE.

RESULTS

A total of 698 patients with SE were included, with a 30-day mortality of 28.9% (202/698). The mean EMSE value in the entire population was 57.1 (SD = 36.3); it was lower in surviving compared to deceased patients (47.1, SD = 31.7 vs. 81.9, SD = 34.8; p < .001). The EMSE was accurate in predicting 30-day mortality, with an AUC ROC of .782 (95% CI = .747-.816). Etiology was the most relevant predictor, followed by age, electroencephalogram (EEG), and EMSE comorbidity group B. The decision tree analysis using EMSE variables correctly predicted the risk of mortality in 77.9% of cases; the prediction was accurate in 85.7% of surviving and in 58.9% of deceased patients within 30 days after SE. In nonhypoxic SE, the most relevant predictor was age, followed by EEG, and EMSE comorbidity group B; the prediction was correct in 78.9% of all cases (89.6% in survivors and 46.1% in nonsurvivors).

SIGNIFICANCE

This validation study using a machine learning analysis shows that the EMSE is a valuable prognostic tool, and appears particularly accurate and effective in identifying patients with 30-day survival, whereas its performance in predicting 30-day mortality is lower and needs to be further improved.

Reponse of second-line treatment in focal status epilepticus: A tertiary hospital experience

OBJECTIVE

To investigate the response to various antiseizure medications (ASMs) in the treatment of focal status epilepticus (SE) in the established phase, and the effect of administering several ASMs prior to sedation.

METHODS

All SE cases in patients aged > 16 years treated with non-BZDs ASMs were prospectively collected in our centre from February 2011 to April 2019. In total, 281 episodes were analysed.

RESULTS

Median age at SE onset was 65.1 years; 47 % were focal motor and 53 % focal non-motor episodes. SE cessation was achieved in 79 % episodes with second-line drugs, whereas a third line (anesthetics) was required in 47 episodes. SE cessation was achieved in only 27 % with the first ASM, 48 % with the second, and 51 % with the third. Prompt resolution of the SE episode with a first or second ASM was associated with a better outcome than episodes requiring a larger number of drugs (p = 0.024). The first option in our sample was levetiracetam in 70 % of cases. Among the total of non-responding SE cases treated with levetiracetam as the first ASM option, 107 were subsequently given lacosamide (seizure cessation in 53.3 %) and 34 valproic acid (seizure cessation in 29.4 %) (p = 0.015).

CONCLUSION

Our findings further support the notion that early termination of SE with a first or second ASM confers a better functional outcome. The large difference in response between the first ASM and consecutive ones suggests that the sum of different ASMs might be the key to resolving focal SE.

Development and Validation of a Nomogram Based on the Epidemiology-Based Mortality Score in Status Epilepticus (EMSE) Parameters to Predict 30-day Mortality in Status Epilepticus

BACKGROUND

To develop a nomogram using the parameters of the Epidemiology-Based Mortality Score in Status Epilepticus (EMSE) and to evaluate its accuracy compared with the EMSE alone in the prediction of 30-day mortality in patients with status epilepticus (SE).

METHODS

We included a cohort of patients with SE aged ≥ 21 years admitted from 2013 to 2021. Regression coefficients from the multivariable logistic regression model were used to generate a nomogram predicting the risk of 30-day mortality. Discrimination of the nomogram was evaluated using the area under the receiver operating characteristic curve (AUCROC) with 95% confidence interval. Internal validation was performed by bootstrap resampling.

RESULTS

Among 698 patients with SE, the 30-day mortality rate was 28.9% (202 of 698). On the multivariable analysis, all EMSE parameters (except for the comorbidity group including metastatic solid tumor or AIDS) were associated with a significantly higher risk of 30-day mortality and were included in the nomogram. The discriminatory capability of the nomogram with bootstrap resampling (5000 resamples) had an AUCROC of 0.830 (95% confidence interval 0.798-0.862). Conversely, the AUCROC of the EMSE was 0.777 (95% confidence interval 0.742-0.813). Thus, the probability that a patient who died within 30 days from SE had a higher score than a patient who survived was 83%, indicating good discriminatory power of the nomogram. Conversely, the risk predicted using the EMSE alone was 77%. The nomogram was well calibrated.

CONCLUSIONS

A nomogram based on EMSE parameters appears superior to the EMSE in predicting the risk of 30-day mortality after SE. The discrimination and calibration of the nomogram shows a better predictive accuracy than the EMSE alone.

Management of Status Epilepticus, Refractory Status Epilepticus, and Super-refractory Status Epilepticus

PURPOSE OF REVIEW

Status epilepticus is a serious condition caused by disorders and diseases that affect the central nervous system. In status epilepticus, hypersynchronous epileptic activity lasts longer than the usual duration of isolated self-limited seizures (time t1), which causes neuronal damage or alteration of neuronal networks at a certain time point (time t2), depending on the type of and duration of status epilepticus. The successful management of status epilepticus includes both the early termination of seizure activity and the earliest possible identification of a causative etiology, which may require independent acute treatment. In nonconvulsive status epilepticus, patients present only with subtle clinical signs or even without any visible clinical manifestations. In these cases, EEG allows for the assessment of cerebral function and identification of patterns in need of urgent treatment.

RECENT FINDINGS

In 2015, the International League Against Epilepsy proposed a new definition and classification of status epilepticus, encompassing four axes: symptomatology, etiology, EEG, and age. Various validation studies determined the practical usefulness of EEG criteria to identify nonconvulsive status epilepticus. The American Clinical Neurophysiology Society has incorporated these criteria into their most recent critical care EEG terminology in 2021. Etiology, age, symptomatology, and the metabolic demand associated with an increasing duration of status epilepticus are the most important determinants of prognosis. The consequences of status epilepticus can be visualized in vivo by MRI studies.

SUMMARY

The current knowledge about status epilepticus allows for a more reliable diagnosis, earlier treatment, and improved cerebral imaging of its consequences. Outcome prediction is a soft tool for estimating the need for intensive care resources.

Unraveling the enigma of new-onset refractory status epilepticus: a systematic review of aetiologies

BACKGROUND AND PURPOSE

New-onset refractory status epilepticus (NORSE) is a clinical presentation, neither a specific diagnosis nor a clinical entity. It refers to a patient without active epilepsy or other pre-existing relevant neurological disorder, with a NORSE without a clear acute or active structural, toxic or metabolic cause. This study reviews the currently available evidence about the aetiology of patients presenting with NORSE and NORSE-related conditions.

METHODS

A systematic search was carried out for clinical trials, observational studies, case series and case reports including patients who presented with NORSE, febrile-infection-related epilepsy syndrome or the infantile hemiconvulsion-hemiplegia and epilepsy syndrome.

RESULTS

Four hundred and fifty records were initially identified, of which 197 were included in the review. The selected studies were retrospective case-control (n = 11), case series (n = 83) and case reports (n = 103) and overall described 1334 patients both of paediatric and adult age. Aetiology remains unexplained in about half of the cases, representing the so-called 'cryptogenic NORSE'. Amongst adult patients without cryptogenic NORSE, the most often identified cause is autoimmune encephalitis, either non-paraneoplastic or paraneoplastic. Infections are the prevalent aetiology of paediatric non-cryptogenic NORSE. Genetic and congenital disorders can have a causative role in NORSE, and toxic, vascular and degenerative conditions have also been described.

CONCLUSIONS

Far from being a unitary condition, NORSE is a heterogeneous and clinically challenging presentation. The development and dissemination of protocols and guidelines to standardize diagnostic work-up and guide therapeutic approaches should be implemented. Global cooperation and multicentre research represent priorities to improve the understanding of NORSE.

Clinical phenotypes within nonconvulsive status epilepticus

The study aimed to identify distinct phenotypes within nonconvulsive status epilepticus (NCSE). Consecutive episodes of NCSE in patients at least 14 years old were included. The level of consciousness was assessed through the Glasgow Coma Scale (GCS). Etiology of NCSE was defined as symptomatic (acute, remote, progressive) or unknown. Electroencephalographic (EEG) recordings were searched for lateralized periodic discharges (LPDs), generalized sharply and/or triphasic periodic potentials (GPDs), and spontaneous burst suppression (BS). According to treatment response, NCSE was classified as responsive, refractory, or superrefractory. Average linkage hierarchical cluster analysis was performed with Pearson correlation as similarity measure. Two hundred twenty‐nine episodes of NCSE were included. Three clusters were identified. The first cluster linked GCS score 3–8, presence of spontaneous BS on EEG, acute symptomatic etiology, and treatment superrefractoriness. The second cluster gathered GCS score 9–12, presence of LPDs or GPDs on EEG, unknown etiology, and treatment refractoriness. The third cluster associated GCS score 13–15, absence of LPDs, GPDs, and spontaneous BS on EEG, and progressive and remote symptomatic etiology with treatment responsiveness. Phenotyping the heterogeneity of NCSE into electroclinical clusters can contribute to understanding correlations between pathologic and clinical domains, assessing the intrinsic severity of NCSE episodes, and estimating the likelihood of treatment responsiveness.