New onset refractory status epilepticus: Long‐term outcomes beyond seizures

Abstract

We propose and prioritize important outcome domains that should be considered for future research investigating long‐term outcomes (LTO) after new onset refractory status epilepticus (NORSE). The study was led by the international NORSE Institute LTO Working Group. First, literature describing the LTO of NORSE survivors was identified using a PubMed search and summarized to identify knowledge gaps. Subsequently, a consensus‐building process was performed to prioritize and rank important LTO domains for further research. The prioritization of LTO domains was qualitative, enabling the expert panel to generate ideas, share opinions, and provide reasons for the rankings. A second round took place to allow expansion and agreement regarding specific details for each domain. Outcomes were classified into eight main domains: (1) Function: Neuropsychological, Neurological (other than seizures), and Psychiatric (mood and behavior); (2) Quality of Life; (3) Epilepsy; (4) Nonneurological (medical); (5) Social; (6) Caregiver Burden; (7) Long‐Term Mortality; and (8) Health Care System Impact. In addition, the working group suggested obtaining outcome measures for each domain at 6 months and 1 year after discharge and annually thereafter until stability has been reached. There are no currently established time frames set for when LTO in NORSE begin or plateau, and previously there existed no consensus regarding which LTO should be considered. This consensus process identifies and recommends NORSE LTO domains that should be considered in future research studies to provide more consistent results that can be compared between studies. Survivors of NORSE should be evaluated serially and at fixed points over time to maximize our understanding of the recovery trajectory for all LTO domains. Establishing reliable and standardized data describing LTO (beyond seizures) after NORSE will support discussions with families during the acute stages, prognostication, the development of targeted management strategies for survivors, and future comparative research globally helping to identify biomarkers that may predict LTO.

Key points.

• There is considerable heterogeneity in LTO assessment methods and their crude conceptualization of favorable or unfavorable in NORSE survivors.

• A consensus proposes eight main LTO after NORSE: Functioning (Neuropsychological, Neurological, and Psychiatric), QoL, Epilepsy, Medical, Social, Caregiver Burden, Mortality, and Health Care System Impact.

• Given that the trajectory of recovery is not well understood, serial measurement of LTO (6 months and 1 year after discharge and then annually) should be used until stability is established.

• Obtaining reliable data describing LTO post‐NORSE will support prognostication and development of targeted management strategies to optimize recovery.

1. INTRODUCTION

Refractory status epilepticus (RSE) and super‐RSE (SRSE) are life‐threatening neurological conditions, associated with high mortality and morbidity. ¹ , ² , ³ , ⁴ When RSE/SRSE occurs in previously healthy individuals without an acute or active structural, toxic, or metabolic cause, it constitutes a clinical presentation (rather than a specific disease) termed new onset refractory status epilepticus (NORSE). ⁵ The use of the term NORSE facilitates its recognition and enables research aimed at guiding and improving management inclusive of prognostic counseling about this potentially devastating syndrome. ⁶ Febrile infection‐related epilepsy syndrome (FIRES) is a subtype of NORSE that is preceded by a febrile illness starting at least 24 h prior to status epilepticus onset. ⁵ Both syndromes may occur in children or adults.

More than 50% of NORSE patients are cryptogenic despite an extensive, albeit variable, diagnostic workup. ⁷ For the remaining patients, paraneoplastic or nonparaneoplastic autoimmune encephalitis (AE) is the most common etiology in adults, whereas identifiable preceding or concomitant infections are more common in the pediatric population. ⁸ Pathogenic genetic variants may have a causative or contributory role in a subset of patients, especially in children, but timely genetic testing is often not available or not pursued. ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹

Outcomes after NORSE vary widely. The in‐hospital mortality rate of NORSE is between 10% and 42%. Variability may depend on etiology, age at presentation, duration of RSE, acute medical complications, and underlying comorbidities as well as existing practice variation for withdrawing life‐sustaining therapies across different centers and countries. ⁹ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷

Ongoing drug‐resistant epilepsy and cognitive changes have been described across the continuum of care including hospital discharge and at follow‐up of variable duration. Unfortunately, there is considerable heterogeneity in outcome assessment tools used, including specific methods, the timing of measurement, and the crude conceptualization of unfavorable and favorable clinical outcomes. A recent systematic review demonstrated that the assessment of the long‐term outcomes (LTO) after NORSE is mainly focused on epilepsy. ¹⁸ However, the sequelae after NORSE include much broader neurological and psychiatric symptoms with social and functioning deficits that greatly impact quality of life (QoL). ¹⁹ , ²⁰ Beyond its direct impact on patients, NORSE often causes a remarkable burden on caregivers and the health care system.

This article proposes priorities and future directions for research investigating LTO after NORSE based on expert consensus. We aim to summarize the existing literature describing LTO of NORSE survivors and to identify gaps and opportunities for future research. Identifying, studying, and closing the divide for outcomes other than epilepsy will facilitate the more holistic support of patients, thus enhancing our ability to monitor their recovery in a more standardized manner. Holistically investigating outcomes will also help to develop and advocate for more evidence‐based rehabilitation that embraces an individualized multimodal approach. In addition, because NORSE is a rare syndrome, the standardization of outcomes measurement will enable future comparative research globally and help us identify biomarkers that are predictive of outcomes.

2. MATERIALS AND METHODS

The study was led by members of the NORSE Institute LTO Working Group (P.H.E., M.T.G., L.J.H., N.W.W., G.K., L.S., A.v.B., H.A.H., O.T., M.A.K., S.A.V., K.E., T.E.G.), which constitutes a group of researchers who meet four times annually and aim to stimulate and integrate research specifically related to NORSE LTO.

We used a consensus‐building process to identify research priorities for LTO after NORSE. First, we prepared a list of important domains based on gaps in the current literature to be discussed, prioritized, and commented upon by a broader group of experts. To review and summarize the current literature describing LTO of NORSE survivors, references were identified using a PubMed search through June 2024. The search terms used combined the following: “new‐onset refractory epilepsy” OR “NORSE” OR “febrile infection‐related epilepsy syndrome”, OR “FIRES” OR “devastating epileptic encephalopathy school‐aged children” AND “neuropsych* outcomes” OR “neurological outcomes” OR “psychiatric outcomes” OR “quality of life” OR “epilepsy” OR “medical outcomes” OR “social outcomes” OR “caregiver burden” OR “long‐term mortality” OR “health care system”. Articles were also identified through searches of the authors' files. The final reference list was generated based on quality and relevance to the scope of this review. The literature review is summarized in Table 1.

TABLE 1.

Summary of current evidence regarding long‐term outcomes after NORSE/FIRES. ^a

First author (year), country	Type of study, b population (N)	Time of FU, median (range)	Neuropsychological outcomes	Neurological outcomes	Psychiatric outcomes	Epilepsy/seizures or status of treatment	Social outcomes
Costello (2024) c	Multicenter, adults (48)	23 months (11–39 months)	79% (31/39) significant cognitive impairment	n = 39; mRS at 6 months: 0: 4% 1–2: 38% 3–5: 44%	54% (21/39) psychiatric issues, mainly anxiety or depression (SR)	n = 39; 100% on ASM; 21% seizure‐free for 12 months at last FU; 79% abnormal EEG on EMU admission, 45% multifocal or BI‐IED	23% (7/30) resumed work 71% (5/7) able to return to education
Stretti (2024), Switzerland	Single center, adults (25)	24 months (17–33 months)	N.A.	n = 16; median mRS at last FU = 2 (1.75–3)	N.A.	N.A.	N.A.
Alaskar (2024), Saudi Arabia	Single center, >14 years old (24)	65 months	N.A.	n = 24; 20% mRS = 0–2; median mRS at last FU = 3 (2–4)	N.A.	n = 18; 88% epilepsy; 43% (7/16) DRE; median ASMs = 3 (2–4 ASM); 26% CI	N.A.
Jimenez (2024)	Multicenter, >13 years old, (64; 92% adults)	Up to 24 months	N.A.	n = 54; median GOS‐E = 5 (moderate disability; IQR = 1–7)	N.A.	n = 42; 57% epilepsy (defined as 1 unprovoked seizure after NORSE)	N.A.
Shi (2023), China	Single center, >16 years old (11)	20 months (12–112 months)	N.A.	6/6 mRS = 0–3	N.A.	5/6 recurrent seizures, 100% DRE; 2/6 remote recurrent SE	N.A.
Shrestha (2023), USA	Multicenter, children (6)	40 months (35–51 months)	None returned to neurocognitive baseline; 83% low IQ	PCPC = 3: 66%; PCPC = 4: 33%	50% psychosocial difficulties and hyperactivity	100% epilepsy, all DRE; 16% VNS; 16% VNS + RNS	83% returned to school; 40% special education
Werbaneth (2022), USA	Single center, adults (21)	4.5 years C‐NORSE 2.5 years NORSE post‐AE	67% (8/12) cognitive impairment d	mRS = 0–2 C‐NORSE: 64% (7/11) at 6 months, 82% (9/11) at 12 months NORSE post‐AE: 100% (2/2) at 12 months	C‐NORSE: 50% (6/12) depression; 1 ADHD; 1 PNES NORSE post‐AE: none	n = 15; 73% recurrent seizures; 80% on ASM	23% (3/13) were driving at last FU
Wu (2021), China	Single center, children (92)	6 months	N.A.	74% (53/71) good outcome (able to live independently or returned to baseline)	N.A.	n = 71; 38% epilepsy; 39% (15/38) DRE	N.A.
Matthews (2020), USA	Single center, adults (26)	9 months (2–22 months)	N.A.	mRS = 0–3: 71% (12/17)	N.A.	n = 17; 88% long‐term ASM	N.A.
Nass (2020), Germany	Single center, >16 years old (6)	3.5 years (1–6 months)	66% (4/6) memory/planning problem; 50% (3/6) attentional problems; 33% (2/6) impaired writing; 16% (1/6) impaired reading e	mRS = 1–2: 66% (4/6); mRS = 3: 33% (2/6)	33% (2/6) social behavior difficulties	n = 6; 100% long‐term ASM	All previous employed (n = 3) were incapacitated to work 1/6 achieved a high school diploma with sickness‐related delay of 2 years
Aurangzeb (2020), UK	Single center, adults (7)	Unknown	Significant memory and/or cognitive impairment (3/3)	mRS = 3–4: 42% (3/3)	Depression: 1/3 Behavioral problems: 1/3	n = 3; 100 % epilepsy; recurrence of SE in 1/3	N.A.
Gugger (2020), USA	Single center, adults (20)	14 months (2–42 months)	N.A.	mRS = 0–2: 40%	N.A.	n = 15; epilepsy 80%; 75% DRE; number of ASMs = 3 (2–4 ASM); CI: 27%; KD: 20%; NS: 13%	N.A.
Lam (2019), China	Single center, children (25)	9 months–15 years	n = 18; 61 % cognitive deficits (38% mild, 11% moderate,11% severe); 11% memory problems	mRS = 0: 27%; mRS = 1–3: 44%; mRS = 4–6: 27% (from 15%, 15 % and 35% at discharge, respectively)	n = 18; 5% ADHD; 5% emotion lability	n = 18; epilepsy 100%; 85% DRE	N.A.
Husari (2019), USA	Single center, <21 years old (29)	5 years (.5–13 years)	n = 23; WISC‐R: 26% normal; 26% learning disabilities and mild–moderate mental retardation; 48% severe mental retardation or vegetative status	N.A.	N.A.	n = 23; 87% epilepsy; median ASMs = 3 (1–5 ASM)	N.A.
Meletti (2017), Italy	Single center, >6 years old (12)	Unknown	n = 23; 65% cognitive deficits (40% severe) f	N.A.	n = 23; 32% behavioral alterations (impulsivity, psychosis, aggression, anxiety) f	n = 23; 78% epilepsy f	N.A.
Iizuka (2017), Japan	Single center, adults (11)	11 months (6–111 months)	N.A.	mRS = 3–5: 73% (8/11)	N.A.	N.A.
Jayalakshmi (2016), India	Single center, adults and children (33)	6 months	N.A.	n = 18; GOS = 5 (low disability): 72% (12/18)	N.A.	N.A.
Gaspar (2015)	Multicenter, adults (130)	9 months (4–19 months)	N.A.	n = 63; mRS = 0–1: 41%; mRS = 2–3: 38%; mRS = 4–5: 16%	N.A.	n = 63; 37% recurrent seizures; 55% not seizures but treated
Howell (2011), Australia	Single center, children (7)	9 years (5–17 months)	n = 6; 66% moderate to severe intellectual impairment; 33% borderline (memory, executive, and attention problems)	n = 6; 66% aggression and impulsivity; 50% taking antidepressants or typical antipsychotics	N.A.	n = 6; 100% DRE	N.A.
Kramer (2011)	Multicenter, children (77) g	At least 1 year	18% normal with or without attention deficit/learning disabilities; 16% borderline; 14%, 24%, and 12% mild, moderate, and severe mental retardation, respectively; 16% vegetative state	N.A.	Severe peripheral neuropathy and ataxia (unknown number)	n = 68; 97 % epilepsy; 93% DRE	N.A.
Caraballo (2009), Argentina	Single center, children (14) h	6.5 years (1–15 years)	2 patients “good outcomes”; “most patients” had mental retardation	Behavioral disturbances (number unclear)	N.A.	100% epilepsy; 83% DRE	N.A.
Costello (2009), USA	Single center, adults (6)	1.5–11 years	n = 5; impaired memory: 40% severe and 20% moderate; 20% moderately impaired executive skills; 20% good recovery (18 months)	n = 5; 40% impaired and dependent; 40% impaired but independent; 20% good	N.A.	n = 5; 60% DRE; 20% not seizures but treated; 20% seizure‐free not treated	n = 5; 80% returned to work; 20% recovered driving license

Note: There are limited data for quality of life, nonneurological (medical), caregiver burden, long‐term mortality, and health care outcomes. These were not included in the table. Refer to the text for further comments.

Abbreviations: ADHD, attention‐deficit/hyperactivity disorder; AE, autoimmune encephalitis; ASM, antiseizure medication; BI‐IED, bilateral independent interictal epileptiform discharges; CI, chronic immunotherapy; C‐NORSE, cryptogenic NORSE; DRE, drug‐resistant epilepsy; EEG, electroencephalogram; EMU, epilepsy monitoring unit; FIRES, febrile infection‐related epilepsy syndrome; FU, follow‐up; GOS, Glasgow Outcome Scale; GOS‐E, GOS‐Extended; IQ, intelligence quotient; IQR, interquartile range; KD, ketogenic diet; mRS, modified Rankin Scale; N.A., not applicable; NORSE, new onset refractory status epilepticus; NS, neurostimulation; PCPC, Pediatric Cerebral Performance Category Scale; PNES, psychogenic nonepileptic seizures; RNS, responsive neurostimulation; SE, status epilepticus; SR, self‐reported; VNS, vagus nerve stimulator; WISC‐R, Wechsler Intelligence Scale for Children–Revised.

^a Includes studies with at least 6 months of follow‐up and more than five patients.

^b All studies were retrospective.

^c Multicenter across Critical Care EEG Monitoring Research Consortium.

^d Formal neuropsychological assessment for eight patients; Mini‐Mental State Examination for four patients; Montreal Cognitive Assessment for four patients.

^e Scale used Scores of Independence for Neurologic and Geriatric Rehabilitation.

^f The study included 12 original cases + literature review (19 cases).

^g This study included six new cases and 71 previously reported cases.

^h This study was not included in Kramer et al. (2011).

Additionally, a broader group of experts were invited to participate in a NORSE LTO meeting as part of the NORSE Roundtable Meeting held during the 2023 American Epilepsy Society (AES) annual meeting (December 3, 2023, Orlando, Florida, USA). This meeting (led by K.E. and T.E.G.) aimed to prioritize and rank the most important LTO domains for future studies. Experts were invited to participate based on evidence of extensive clinical experience managing NORSE, ongoing scholarly activity related to NORSE, and a strong track record of successful research collaborations. Two participants from the NORSE Institute LTO Working Group also participated in the NORSE roundtable meeting (A.v.B., N.W.W.). The expert panel was provided with the initial list of domains and instructed to rank them in order of importance. Furthermore, they were asked to identify similar or disparate domains and to provide a qualitative description of each domain including specific areas for further research. Groups could reject domains or suggest entirely new domains for priority. They completed this first task by working in groups of two to four participants. Each group's priority rankings for the domains were compared and discussed within the broader expert panel. The results of this session were documented. This first domain prioritization round was qualitative in nature, enabling the expert panel to generate ideas, share opinions, and provide reasons for the rankings. Twenty‐three participants attended and contributed to the consensus process during the 2023 roundtable in Orlando, Florida. The participants were from North America (United States, Canada, and Mexico), Europe (Belgium, Italy, France, Germany, Sweden), Korea, and New Zealand and included adult and pediatric neurologists with experience managing patients with NORSE, a patient advocate, and a neuropsychologist. Further invited experts were unable to attend the roundtable session in person but were able to participate in the second round of consensus generation. Consensus domain rankings were determined based on cumulative rankings from 1 to 9 (1 = highest priority) among all groups.

In the second round, there were 32 respondents who agreed on the priority domains and their respective rankings, but further comments were made regarding specific details to be considered for each domain. These were addressed through recurrent virtual meetings. The response rate of the second round was 11 of 16 AES participants and 11 of 16 participating NORSE LTO group members. Finally, based on feedback and subsequent discussions, a final ranking of the important LTO domains was circulated electronically to a total of 32 panelists, including the AES panelists and the LTO working group members, for individual comment in January 2024. Anonymity was not maintained during the process, because the first round occurred in person and the second round was completed using email. This approach, however, allowed the clarification of multiple viewpoints, thereby strengthening the diversity of opinion.

3. RESULTS

The outcomes considered in this review are patient‐centered, but we also considered the impact of NORSE on caregivers and health care systems. No proposed priority domains were rejected during the consensus process. Based on consensus from the first and second round, outcomes were classified into eight main domains (Table 2): (1) Function: Neuropsychological, Neurological (other than seizures), and Psychiatric (mood and behavior); (2) QoL; (3) Epilepsy; (4) Nonneurological (medical); (5) Social; (6) Caregiver Burden; (7) Long‐Term Mortality; and (8) Health Care System Impact. Although there is likely overlap between domains, and many, such as Social or QoL, may contribute to the broader category of “function‐related outcomes,” the working group recommended maintaining these as separate domains to underscore their importance. Similarly, epilepsy was considered separately from other neurological outcomes.

TABLE 2.

Summary of outcomes and recommendations for the eight main domains proposed.

Domain/subdomain	Outcome/variable	Recommendations
I. Functional
I.1 Neuropsychological	‐ IQ ‐ Memory ‐ Language ‐ Executive ‐ Visuospatial	‐ Serial comprehensive neuropsychological assessments at 6 months and 1 year after discharge and annually thereafter until stability. More frequent assessments may be needed for patients. ‐ The choice of the test is driven by patient age, local preferences, and resources available. Optimal tests for in‐depth and abbreviated assessment need further exploration. a ‐ When appropriate, instruments must be chosen based on etiology of NORSE b and considering specific necessities of the patients c (e.g., hearing, motor, visual impairments). ‐ When the application of a formal test is not possible, the cognitive domains must be assessed in a highly structured way. ‐ For infants and toddlers, assessment must consider the developmental curves. ‐ Regardless of the tests, the raw score should be documented to allow later retrospective analysis.
I.2 Neurological	‐ General performance (e.g., mRS, GOS, GOS‐E, GIS, PCPC, CGI, GMCS) ‐ Level of consciousness ‐ Motor deficits ‐ Movement disorder ‐ Sensory deficits ‐ Ataxia ‐ Visual defects ‐ Sleep disturbances	‐ Serial assessments at 6 months and 1 year after discharge and annually thereafter until stability. ‐ Current scales used for general performance do not provide comprehensive assessment of deficits. ‐ Avoid dichotomizing results as poor or good outcomes and indicate the raw score to allow later retrospective analysis.
I.3 Psychiatric	‐ Depression ‐ Anxiety ‐ Obsessive–compulsive disorder ‐ Bipolar disorder ‐ Panic disorder ‐ Learning difficulties ‐ Psychosis ‐ Behavior dysregulation ‐ Posttraumatic stress disorder ‐ Substance abuse ‐ Suicidality	‐ Serial screening at 6 months and 1 year after discharge and annually thereafter until stability. ‐ Further reference to psychiatric specialist as needed. ‐ Several available tests exist, but there is no specific evidence supporting their use in NORSE. d
II. Quality of life	‐ QOLIE‐10 ‐ QOLCE‐16 (pediatrics)	‐ Assessment at 6 months and 1 year after discharge and annually thereafter until stability. ‐ Minimal data to support recommendations.
III. Epilepsy	‐ Diagnosis ‐ Time of the first unprovoked seizure ‐ Classification: focal (include regional or multifocal), generalized, and combined ‐ Burden of seizures: cluster, recurrence of status epilepticus, seizure‐related injury ‐ ASM burden (number and defined daily dose) ‐Response to treatment (seizure freedom, seizure reduction) and side effects: pharmacoresponse (ASM, steroids, IVIg, etc.) versus MRE: surgery (resection vs. palliative); neuromodulation: VNS, DBS, RNS, dual neuromodulation; ketogenic diet therapy ‐ EEG findings: IED, ictal findings, abnormal background e ‐ MRI: mesial temporal sclerosis, atrophy (volumetric analysis), signal abnormality (compare with the acute stage) ‐ PET/SPECT for presurgical evaluation	‐ Assessment at 6 months and 1 year after discharge and annually thereafter until stability. ‐ The diagnosis of epilepsy post‐NORSE remains retrospective and is likely influenced by the etiology. ‐ Avoid overdiagnosis of epilepsy. Report status of seizures and ASMs rather than specific diagnosis of epilepsy when the latter is unclear. ‐ Comparing the findings with the acute stages is recommended. ‐ EMU admission for presurgical investigations in refractory patients.
IV. Nonneurological (medical)	‐ Renal (kidney injury), gastrointestinal (bleeding, perforation, pancreatitis, ischemia, ileus), cardiac (arrhythmia, infarct, Takotsubo cardiomyopathy), respiratory (ARDS, pulmonary embolism, tracheostomy, aspiration), or hematologic (deep venous thrombosis) ‐ Growth in children	‐ Acute medical complications are not uncommon during RSE/SRSE, but their occurrence in the long term is not reported. ‐ Follow‐up by the corresponding specialties if acute complications occurred. Timeline depends on the type and severity of complications. ‐ Involvement of the patient's primary care provider is strongly recommended.
V. Social	‐ Activities of daily living ‐ Work status: pre‐ and post‐NORSE; need for employment accommodations ‐ Educational status: pre‐ and post‐NORSE; academic performance, final academic attainment, and special needs ‐ Driving restrictions ‐ Degree of community support	‐ Assessment at 6 months and 1 year after discharge and annually thereafter until stability.
VI. Caregiver burden	‐ BSFC and the Zarit caregiver burden inventory ‐ Work status ‐ New diagnosis of psychiatric comorbidities ‐ Change of economic status	‐ Assessment at 6 months and 1 year after discharge and annually thereafter until stability. ‐ Offering psychological, social, and financial supports may benefit caregivers and patients.
VII. Mortality	‐ Long‐term mortality	‐ Assessment at 6 months and 1 year after discharge and annually thereafter. ‐ Focus on SUDEP and seizure‐related deaths versus medical complications.
VIII. Health care	‐ Long‐term costs ‐ Emergency department visits ‐ Hospital readmission ‐ Outpatient visits (medical services; speech, occupational, and psychological services)	‐ Assessment at 6 months and 1 year after discharge and annually thereafter. ‐ Include costs for continuing care and acute decompensations (e.g., seizure recurrence).

Abbreviations: ARDS, acute respiratory distress syndrome; ASM, antiseizure medications; BSFC, Burden Scale for Family Caregivers; CGI, Clinical Global Impression of Improvement; DBS, deep brain stimulator; EEG, electroencephalogram; EMU, epilepsy monitoring unit; GIS, Global Impression Scale; GMCS, Gross Motor Function Classification System; GOS, Glasgow Outcome Scale; GOS‐E, GOS‐Extended; IED, interictal epileptiform discharges; IQ, intelligence quotient; IVIg, intravenous immunoglobulin; MRE, medically refractory epilepsy; MRI, magnetic resonance imaging; mRS, modified Rankin Scale; NORSE, new onset refractory status epilepticus; PCPC, Pediatric Cerebral Performance Category Scale; PET, positron emission tomography; QOLCE‐16, Quality of Life in Epilepsy Inventory–16; RNS, responsive neurostimulation; RSE, refractory status epilepticus; SPECT, single photon emission computed tomography; SRSE, super‐RSE; SUDEP, sudden unexpected death in epilepsy; VNS, vagus nerve stimulator.

^a Cognitive batteries in adults: Wechsler Adult Intelligence Scale–Fourth Edition (high reliability, Cronbach alpha > .90, and strong construct validity), Wechsler Memory Scale–Revised (high reliability, Cronbach alpha > .85, and well‐established validity), Montreal Cognitive Assessment (good reliability, Cronbach alpha ~ .83, and strong construct validity), Mini‐Mental State Examination (high test–retest and interrater reliability, good construct validity). Cognitive batteries in children: Wechsler Intelligence Scale for Children–Revised (reliability: Cronbach alpha > .90, validity: strong construct validity), Terman Merrill, European Oral Language Battery, Signoret Memory Battery, and Rey–Osterrieth Complex Figure test.

^b Repeatable Battery for Neuropsychological Status (high reliability, Cronbach alpha > .80, and strong validity) has been effective in patients with autoimmune encephalitis and is recommended for those with NORSE, especially with autoimmune etiology.

^c Leiter International Performance Scale–Third Edition (reliability: Cronbach alpha > .80, validity: strong) is effective for deaf or hard‐of‐hearing and international populations. Differential Abilities Scale–Second Edition (reliability: Cronbach alpha > .85, validity: strong) and Kaufman Assessment Battery for Children–Second Edition (reliability: Cronbach alpha > .85, validity: strong) may yield a more valid full‐scale IQ for children with heterogeneous outcomes. In children with severe cognitive or behavioral deficits, the assessment should also include parent/teacher reports (e.g., Vineland Adaptative Behavior Scales–3).

^d Beck Depression Inventory, Hospital Anxiety and Depression Scale, Generalized Anxiety Disorder 7‐Item, and Patient Health Questionnaire; the Neurological Disorders Depression Inventory in Epilepsy for selected cases with epilepsy. In addition, in children, parent‐based testing may be considered (e.g., the Strengths and Difficulties Questionnaire for behavioral testing).

^e Specify duration and type of EEG: routine, ambulatory, inpatient (presurgical).

3.1. Timing for follow‐up assessment

There have been no established time frames set for what constitutes LTO after NORSE. The trajectory of recovery is not well understood. Outcomes at hospital discharge or 30 days after discharge have been used, with the rationale that hospital discharge reflects clinical stability. ¹⁸ , ¹⁹ , ²¹ There is evidence that a subset of NORSE survivors may demonstrate improvements in their functioning status in the months after discharge from acute care despite long and complex hospital stays. ⁸ , ⁹ , ¹¹ , ¹⁵ , ¹⁷ In addition, there are data showing progressive brain atrophy up to 5 years after the onset of FIRES in children despite a stable clinical course. ²² Furthermore, data from other brain disorders including traumatic brain injury and AE (without NORSE) in adults and acute encephalitis in children suggest that time to neurologic improvement is variable (3–5, 1.5, and .5–1 years after the acute insult, respectively). ²³ , ²⁴ , ²⁵ , ²⁶ As a result, the working group suggested obtaining outcome measures for each domain at 6 months and 1 year after discharge and annually thereafter until stability has been reached. This timeline definition is based on expert consensus and may vary on a case‐by‐case basis. Recovery is a dynamic process, and outcomes may fluctuate, worsen, or improve based on the underlying etiology, treatments, comorbidities, seizure burden, socioeconomic situation, and access to rehabilitation, among other factors; therefore, serial assessments are important. Because the long‐term trajectory of NORSE over years is not well understood, no limits on long‐term follow‐up should be established until more data become available. Following current survivors of NORSE at fixed points over time without restrictions will enhance our understanding of the clinical course for all LTO domains. In addition, LTO assessments offer opportunities to repeat investigations looking for a cause in cryptogenic cases, especially if worsening is noted.

4. DISCUSSION

The eight domains of LTO in NORSE are discussed below.

4.1. Function outcomes

4.1.1. Neuropsychological outcomes

Adult considerations

Most studies published describing neuropsychological outcomes in adults have been single case reports ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² or small case series. ¹⁰ A recent systematic review showed that cognitive outcomes were reported in only 10% of adults with NORSE (22/212), with 27% of reported survivors having moderate to severe cognitive impairment. ¹⁸ However, the cognitive assessment was based on patient or family reports for all cases except a single case where a formal neuropsychological testing showed impaired visual and verbal memory, moderately impaired language function, and low average intelligence quotient (IQ). ³¹ In another retrospective study of 21 adults with NORSE, cognitive outcomes were available for 12 patients, of whom 67% (8/12) had some degree of cognitive impairment at the last follow‐up (median = 4.5 years). The authors used a formal neuropsychological assessment (n = 8), the Mini‐Mental State Examination (n = 4), and the Montreal Cognitive Assessment (n = 4). ¹⁵ A multicenter retrospective study showed that 79% (31/39) had significant enduring self‐reported cognitive difficulty at ≥6 months after discharge. ³³ Furthermore, in a retrospective series of adults with FIRES from Germany, the authors used the Scores of Independence for Neurologic and Geriatric Rehabilitation scale with a mean follow‐up of 3.5 years and found that 83% (5/6) had loss of independence in cognition‐related activities of daily living; 66% (4/6) had impaired memory, planning, or problem‐solving; 50% (3/6) were disabled in terms of attentional demand; 33% (2/6) had impaired writing; and 16% (1/6) had impaired reading comprehension. ³⁴

Pediatric considerations

Cognitive outcomes are more frequently reported in pediatric patients. From the data available of 233 patients, ¹⁸ roughly 80% showed some degree of cognitive impairment, including mild (12%), moderate (17%), severe (14%), and vegetative state (8.7%). In a subset of patients, the degree of impairment was not mentioned. On the other hand, 22% showed complete cognitive recovery. Recently, a multicenter retrospective study of FIRES survivors reported low to extremely low average IQ in five of six patients (median = 40 months). Interestingly, three patients showed regression of full‐scale IQ over time. ³⁵ Other cognitive deficits reported in children included visual–spatial disabilities, ³⁵ memory difficulties, language impairment, ³⁵ , ³⁶ , ³⁷ , ³⁸ executive problems, ³⁸ , ³⁹ , ⁴⁰ and motor speed dysfunction. ³⁵ , ³⁷

Recommendations

Based on the above, neuropsychological outcomes of NORSE in adults and children are underreported and performed using nonstandardized measures that do not provide the level of detail required to develop effective rehabilitative approaches. There is not a specific pattern of cognitive deficits in patients with NORSE; however, different tests have been used yielding heterogeneous data and making their comparison difficult. In addition, the studies carry a high risk of selection bias with respect to who underwent neuropsychological testing and when it was done. Neuropsychological LTO, such as memory, language, attention, executive functions, IQ, and visuospatial functions, should be measured using reliable and valid tools. The selection of cognitive tests is influenced by patient age, local preferences, resource availability (e.g., neuropsychological team), familiarity with the tests, and the situation in which the evaluation takes place (routine outpatient visit vs. neuropsychiatric testing as part of presurgical epilepsy investigation). In addition, a subgroup of survivors will be too impaired to participate in conventional neuropsychological assessments. An experienced neuropsychologist should be involved to consider how to assess their cognitive performance using other methods in a structured way. The Wechsler IQ test is internationally accepted and available in almost every country worldwide, and a psychologist without neuropsychological experience can apply it. When appropriate, specific instruments depending on the etiology must also be considered (e.g., Repeatable Battery for Neuropsychological Status in patients with AE). ⁴¹ Furthermore, in the case of specific impairments (e.g., hearing, motor, visual), there are special needs editions or specific tests. ⁴² For infants and toddlers, the assessment should be done with a standardized developmental test (e.g., Bayley Scales of Infant and Toddler Development) using the most recent version and should consider the developmental curves to represent the dynamics of cognitive development. Regardless of the batteries, the raw score should be documented, to allow subsequent retrospective analysis. Some available tests along with their reliability and validity are listed in Table 2.

Further research is required to devise the optimal approach for both comprehensive and abbreviated cognitive assessments, as well as to determine the optimal timing for these tests. Special attention must be given for patients who continue to decline over time. ³⁵ , ³⁸ Although decline may be multifactorial and due, in part, to initial (RSE) and ongoing (seizures and medication effects) factors, further exploration is needed to determine which factors are modifiable.

4.1.2. Neurologic impairments

Adult considerations

Most studies on LTO in NORSE have reported neurologic functioning outcomes using the modified Rankin Scale (mRS), ⁹ , ¹¹ , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ²⁰ , ³³ , ³⁴ , ⁴³ the Glasgow Outcome Scale (GOS), ¹² , ⁴⁴ and the GOS‐Extended (GOS‐E). ⁴⁵ The two largest multicenter studies of adults with NORSE reported long‐term mRS. In the first one, Gaspard et al. ⁹ found that 41% (26/63) had mRS = 0–1 at median 9 months follow‐up and up to 79% had mRS = 0–3, which was considered a fair or favorable outcome. The second study, by Costello et al., ³³ included 48 patients and reported that 4% had a full recovery (mRS = 0) after 6 months, 38% had mRS = 1–2 (good outcomes), and 44% had mRS = 3–5 (poor outcomes). Similar results have been found by other smaller series. ¹¹ , ¹⁴ , ¹⁶ When authors considered good outcomes as mRS = 0–3, a higher proportion of good outcomes was described: 71% (median follow‐up of 9 months for 17 patients) ¹¹ and 100% (median follow‐up of 20 months for six patients). ¹⁶ Less encouraging results were reported by two other case series, which included nine and 24 patients respectively and reported that roughly 70%–80% of patients had mRS > 2 (median follow‐up of 11 and 20 months, respectively). ²⁰ , ⁴³ Another large multicenter study described a median GOS‐E of 5 (moderate disability) during a follow‐up of up to 24 months in 54 patients older than 13 years. ⁴⁵

Even though most of the studies used mRS to measure functioning outcomes, it is difficult to compare outcomes between studies because of inconsistencies in how the scores are used to report positive versus negative outcomes and timing of follow‐up assessment, diversity of ages, and potential heterogeneous etiologies and management strategies. Furthermore, using mRS as the only scale to report LTO has limitations, especially when the results are dichotomized into two possibilities: “good” and “poor” outcomes. The mRS score does not capture a full clinical picture of deficits, especially for cognitive outcomes. Even patients who achieve a favorable outcome, as defined by a mRS of 2 or lower, may not be unable to return to their previous baseline function because of cognitive deficits or psychiatric symptoms leading to social and QoL impairments. ³⁴

Pediatric considerations

In addition to mRS, GOS, and GOS‐E, other scales have been used on pediatric populations, including the Clinical Global Impressions Scale and the Pediatric Cerebral Performance Category Scale (PCPC). The mRS in children carries the same limitations explored above, and a broad range of results has been described. ¹⁸ Based on PCPC, unfavorable outcomes (score = 3 or more) ranged from 80% ¹⁸ to 100%, ³⁵ with variability and with some possible improvement over time. ³⁵ , ³⁶

Comments and recommendations

Neurologic functioning outcomes and the sequelae of NORSE are far more diverse than what can be measured using current tests (mRS, GOS, GOS‐E, or PCPC). An individualized approach considering the patient and family perspectives is necessary to define "good" or "poor" outcomes. Rather than dichotomizing the results (favorable vs. unfavorable), the raw results should be documented to allow comparison between studies. In addition, other physical neurological sequelae such as weakness, sensory loss, ataxia, movement disorders, visual field deficits, and sleep disturbances should also be considered and reported systematically.

4.1.3. Psychiatric outcomes

Adult considerations

Few studies have formally investigated the presence of new psychiatric symptoms in adult survivors of NORSE. A previous systematic review found that psychiatric symptoms (psychomotor agitation, personality changes, and Capgras syndrome) were reported in only 3% of cases (seven patients from different single case reports). ¹⁸ All except one case report described the timing of onset for psychiatric symptoms to be during the acute phase or soon after NORSE resolved. ⁴⁶ , ⁴⁷ One case report described new onset schizophrenia at 9 months after discharge. ²⁹ More recently, a case series with a follow‐up of at least 6 months described anxiety and depression in up to 54% (21/39) of survivors. ³³ These symptoms were self‐reported by patients. Additionally, psychogenic nonepileptic seizures (PNES) and attention‐deficit/hyperactivity disorder (ADHD) were reported in a small case series after >1 year of follow‐up. ¹⁵

Pediatric considerations

New onset psychiatric disorders were reported in 11.2% of children recovering from NORSE. ¹⁸ The most common psychiatric symptoms reported were behavioral disturbances, including aggression, emotional lability, apathy, and conduct disorder. Other symptoms include ADHD ³⁵ , ³⁶ and depression with suicidal attempts. ³⁹

Comments and recommendations

Psychiatric symptoms have not been measured systematically and are likely underreported. The working group recommends screening for psychiatric symptoms including depression, anxiety, obsessive–compulsive disorder, bipolar disorder, panic disorder, psychosis, behavior dysregulation, ADHD, PNES, posttraumatic stress disorder, substance abuse, and suicidality until further evidence is available to provide more specific recommendations. Some available tests are listed in Table 2; however, no specific evidence for their use in NORSE exists.

4.2. Quality of life

Unfortunately, there are limited data regarding QoL in NORSE survivors. Data from adult patients after SE have shown substantial impairments in QoL (based on Quality of Life in Epilepsy Inventory 31), but this may improve over long‐term follow‐up. ⁴⁸ The NORSE Family Registry is a collaborative project developed by members of the NORSE Institute and is collecting data assessing QoL. ⁴⁹ However, presently there remain limited data regarding QoL adults and children after NORSE.

4.2.1. Recommendations

QoL should be assessed at regular intervals for NORSE survivors. Further research is required to establish the optimal tools for this task.

4.3. Epilepsy

4.3.1. Adult and pediatric considerations

Only 5% of adult and 9% of pediatric survivors of NORSE become seizure‐free without antiseizure medications (ASMs). ¹⁸ Among the patients with persistent seizures, 40%–79% have drug‐resistant epilepsy (DRE). ¹⁴ , ¹⁸ , ³³ Interestingly, although reporting subsequent seizures after NORSE remains the main focus in the literature, there is no consensus on the definition of epilepsy after NORSE, and several studies reported “epilepsy” even when the patients were seizure‐free but continued on ASMs. Because the diagnosis of epilepsy is based on the recurrence risk of seizures driven by the underlying etiology, a specific etiological diagnosis for NORSE should be determined when possible. For patients with AE (the most commonly identified cause of NORSE), RSE and even subsequent seizures are often acute symptomatic events and do not necessarily constitute epilepsy. There is evidence that the likelihood of enduring seizures after appropriate immunotherapy in patients with AE associated with antibodies against cell surface antigens, ⁵⁰ , ⁵¹ , ⁵² , ⁵³ even if they presented with NORSE, is low. ⁴⁶ These patients usually recover slowly compared to patients without NORSE but eventually achieve seizure freedom. Certainly, in a subset of patients with an autoimmune etiology, structural brain changes emerge during the acute phase (demonstrated on brain magnetic resonance imaging (MRI) as nonspecific hyperintensity on fluid‐attenuated inversion recovery/T2 sequences mainly affecting the limbic structures and likely as a consequence of seizure activity), ⁹ , ⁴³ , ⁵⁴ but at times changes may persist chronically (e.g., mesial temporal sclerosis or atrophy) and potentiate ongoing seizures. ²⁰ , ³¹ , ³⁵ , ³⁷ , ³⁹ , ⁵⁵ , ⁵⁶ The International League Against Epilepsy (ILAE) has proposed the term autoimmune‐associated epilepsy to highlight how other nonimmune factors may play a role in post‐AE epilepsy. ⁵⁷ Most recently, Rada and Bien proposed a practical definition for AE‐associated epilepsy, which requires the persistence of seizures for at least 2 years after immunotherapy and the lack of central nervous system signs of acute inflammation (based on neuroimaging and cerebrospinal fluid [CSF] cell count and antibody titers). ⁵⁸ Although this definition is restrictive, the authors emphasized the medical and social consequences of a premature diagnosis of epilepsy. The diagnosis of epilepsy after NORSE should be based on the etiology; however, there is lack of consensus (e.g., for paraneoplastic or infectious etiologies). Other causes including pathogenic variants identified during genetic testing require individualized considerations. Specific criteria for epilepsy after NORSE are needed and should spare patients from a premature diagnosis of epilepsy.

For patients with cryptogenic NORSE, the situation is even more complex. A subset of patients may continue having seizures without a latent period after the RSE/SRSE, especially after the FIRES subtype in children. ³⁹ , ⁵⁹

Furthermore, even though seizures are very commonly reported in survivors of NORSE, details about the seizure and epilepsy classification are lacking. This is important because seizure and epilepsy classification drive the therapeutic approach. Based on current evidence, a high proportion of NORSE survivors have multifocal or regional epilepsy and are not candidates for traditional epilepsy surgery. ³³ Treatment options for these patients include different ASMs (usually polytherapy), neuromodulation, ketogenic diet therapy, or trials of immunotherapy, with variable rate of response but often without obtaining seizure freedom. ⁶⁰ , ⁶¹ , ⁶²

4.3.2. Comments and recommendations

It is important to define epilepsy after NORSE, given the impact of such a diagnosis for ongoing management, but also because of social stigma associated with epilepsy. Currently, the diagnosis of epilepsy after NORSE remains retrospective. Epilepsy monitoring unit admission plays a crucial role in enabling better characterization of seizure type in survivors of NORSE. Comparing the electroencephalogram (EEG) and MRI in the chronic stage with the acute phase may shed light on NORSE‐specific pathophysiology of epilepsy. The working group suggests that the following variables should be considered when reporting seizure/epilepsy outcomes: time of the first unprovoked seizure after the acute stage, classification (based on ILAE recommendations), ASM burden (number and defined daily dose of ASM), seizure severity (from patient/caregiver perspective), recurrence of status epilepticus, cluster recurrence during fever, EEG findings (routine, ambulatory, or epilepsy monitoring unit), MRI findings, biology (serum, CSF, microbiome when available and depending on the underlying etiology), positron emission tomography/magnetoencephalography/single photon emission computed tomography results, response to treatment (seizure freedom, seizure reduction), side effects of ASM, epilepsy surgery (resective, neuromodulation), ketogenic diet, and chronic immunotherapy.

4.4. Nonneurological health outcomes

4.4.1. Adult and pediatric considerations

Patients with RSE/SRSE have a high risk of acute medical complications, ⁶³ , ⁶⁴ including renal (acute kidney injury), gastrointestinal (bleeding, perforation, pancreatitis, ischemia, ileus), cardiac (arrhythmia, infarct, Takotsubo cardiomyopathy), respiratory (acute respiratory distress syndrome, pulmonary embolism, tracheostomy, aspiration), and hematologic (deep venous thrombosis), among others. However, little is known about the long‐term impact of these medical issues and complications on morbidity, mortality, and other outcomes such as growth and development in children. Lee and Chi reported irreversible liver failure after FIRES, although the specific cause of this was not elucidated and likely was multifactorial. ⁵⁶

4.4.2. Comments and recommendations

Certainly, these medical complications may affect neurologic function and social outcomes and QoL and should be monitored. The working group recommends following up any medical conditions diagnosed during the acute stage. Further studies regarding the nature of nonneurologic health outcomes and the optimal timing for follow‐up monitoring are needed.

4.5. Social outcomes

4.5.1. Adult and pediatric considerations

Social outcomes, including work status, education, and academic performance, and driving restrictions (depending on local regulations) impact the QoL and level of social functioning and return to community. Few data have been published regarding social outcomes, and existing data are heterogeneous. Resumption of premorbid employment has been documented in 23% (7/30) of patients from a multicenter study with a follow‐up of at least 6 months and in 80% (4/5) from another single‐center study. ¹⁰ , ³³ Nass et al. ³⁴ found that 50% of previously employed patients returned to work after NORSE, whereas 16% needed to change their employment. Most children required special education services. ³⁵ , ³⁷ , ³⁸ , ⁵⁵ Substantially less is known about other social outcomes.

4.5.2. Comments and recommendations

The working group recommends including the following outcomes for future research: activities of daily living; work status (pre‐ and post‐NORSE) and any employment accommodations; educational status (pre and post‐NORSE), including academic performance, final academic attainment, and special needs; driving restrictions; and the degree of community support. Social outcomes should also be assessed clinically, and support should be provided to foster social functioning and return to community activities.

4.6. Caregiver burden

4.6.1. Adult and pediatric considerations

Caring for a family member or loved one with complex medical issues after hospitalization with NORSE leads to emotional, physical, and social distress in caregivers, with an impact on family dynamics, social life, employment, possibly financial status, and QoL. This is potentially worse when the condition comes on suddenly, and neurologic deficits and medical complexity occur suddenly in previously healthy patients, which is the case in NORSE. In most cases, the primary caregiving responsibilities fall on family members who support their loved ones on a path full of uncertainty. There is a paucity of research exploring effects of NORSE on caregivers. Data from individuals with epilepsy suggest that caregiver burden negatively impacts caregivers' health‐related QoL. ⁶⁵ , ⁶⁶ , ⁶⁷ Roughly 50% of caregivers of patients with severe epilepsy syndromes report symptoms of depression. ⁶⁸ The caregiver burden may vary depending on the degree of the disability of the survivor.

4.6.2. Comments and recommendations

Incorporating caregiver assessment and offering psychological, social, and financial support may benefit caregivers and patients, because the physical and mental health of so many NORSE patients is dependent upon their caregivers. Considerations for caregivers may include emotional and physical domains. The Burden Scale for Family Caregivers and the Zarit Caregiver Burden Inventory may be used. These batteries have been used to assess caregiver burden in various neurologic disorders, including epilepsy. ⁶⁹ , ⁷⁰ , ⁷¹

4.7. Mortality

4.7.1. Adult and pediatric considerations

Although in‐hospital mortality has been described to be between 10% and 42%, ¹³ , ¹⁵ the long‐term mortality associated with NORSE is unknown. It is not clear whether the RSE/SRSE itself may contribute to long‐term mortality. ⁷² Costello et al. ³³ reported that 19% (9/48) of patients died after discharge, with 80% of deaths occurring during the first 6 months. Beyond 6 months, long‐term mortality may be influenced by the underlying etiology of NORSE, age, medical comorbidities, and new onset of DRE. Deaths between 6 months and 8 years after NORSE have been reported in case reports, with some deaths attributed to sudden unexpected death in epilepsy (SUDEP) ¹⁴ , ¹⁶ , ⁵⁴ and respiratory failure. ⁵⁶

4.7.2. Comments and recommendations

Further research is necessary to understand whether RSE/SRSE contributes to long‐term mortality. A specific focus on SUDEP is recommended.

4.8. Health care system impact

4.8.1. Adult and pediatric considerations

Annual direct costs for in‐hospital treatment of SE are approximately US$4 billion in the USA and >€200 million in Germany. ⁴⁸ These data cannot be directly extrapolated to NORSE, and there is no evidence available assessing long‐term resource utilization after NORSE.

4.8.2. Comments and recommendations

Most NORSE survivors will continue to require frequent lifelong contact with the health care system. Beyond cost, emergency department visits, hospital readmission, and outpatient visits should be considered as outcomes of interest. Furthermore, inpatient costs may be reflected by documenting the total days of ventilatory support, days in the intensive care unit, and days in the hospital. Finally, there are many outpatient costs for physical, speech, occupational, and psychological therapy required to address LTO. This is an important area to consider in terms of support for patient and families, such as ensuring access to health insurance. Additionally, evaluating the impact to the health care system may help prioritize NORSE research, despite the rarity of the condition.

4.9. Role of NORSE etiology in outcomes

Several factors, including differences in populations studied, symptomatic etiology versus cryptogenic, treatments, medical comorbidities, outcome measurement scales, and variable follow‐up, may contribute to the heterogeneity of reported LTO after NORSE. The etiology of NORSE likely influences outcome, but results to date are conflicting. Better outcomes have been reported for patients with NORSE in association with AE, ¹⁵ , ⁴⁶ and less favorable results have been reported for survivors of cryptogenic NORSE. ²⁰ On the other hand, some studies have found no significant differences when comparing cryptogenic NORSE versus NORSE from an ultimately identified etiology. ⁹ , ⁷³ There are also case reports of cryptogenic NORSE with favorable outcomes. ¹⁵ , ⁷³ Considering that NORSE becomes cryptogenic by exclusion means that its identification may depend on the resources or tests available, and therefore a minimum standard panel of investigations should be proposed to define a case of NORSE as cryptogenic. Patients not meeting this minimum standard would be considered to have NORSE of undetermined etiology (due to incomplete investigations) rather than strictly cryptogenic NORSE. This strategy may help to more effectively differentiate and analyze the true cryptogenic cases that most accurately reflect our incomplete and current knowledge. Reporting outcomes in relation to etiology will further our understanding of NORSE and the long‐term effects of treatment.

4.10. Multidisciplinary approach for NORSE survivors

NORSE may lead to several consequences summarized in the eight domains. The diversity and complexity of the burden of these sequelae require an approach from different fields of health care professionals to provide the best patient care. Although neurologists play a central role during the acute stage of NORSE, during the long‐term follow‐up, collaboration with other specialties is imperative and should include psychiatry, psychology, neuropsychology, radiology, neurosurgery (in the context of epilepsy surgery assessment), and other specialties, depending on the individual's needs (e.g., internal medicine, rheumatology, physical medicine and rehabilitation, and occupational therapy, among others). In addition, it is essential to involve the primary care physician, who will coordinate the team input from a variety of disciplines, helping to reduce gaps between different perspectives in clinical practice. Furthermore, given the social and QoL impairment of the patients and their families, social workers and specialized nurses are very important for providing additional support. A multidisciplinary team approach will allow us to provide better patient care for survivors of NORSE in the longer term.

5. CONCLUSIONS

It is critical to obtain reliable data describing LTO after NORSE. This will support discussions with families during the acute stages, prognostication, and development of targeted management strategies for survivors. Herein, the authors, who have extensive experience managing patients with NORSE, have proposed priority areas for research investigating NORSE LTO. Screening for post‐NORSE functioning (neurological, neuropsychological, and psychiatric), QoL, epilepsy, medical complications, social, long‐term mortality, caregiver burden, and health care system burden would reflect a holistic approach for survivors of NORSE and would inform appropriate interventions to optimize recovery. In addition, standardizing LTO will provide a first step for further research‐oriented clinical outcome studies, allowing the assessment of any intervention during the acute or chronic stages of NORSE.

Although there are not yet data to support how frequently and for how long follow‐up should occur, we recommend serial measurement of LTO at 6 months and 1 year after discharge and then annually until stability. This frequency may be variable based on individual patient and caregiver needs. Current data suggest that the degree of recovery after NORSE is extremely variable and is likely influenced by the etiology, so LTO should be reported considering the etiology. Cryptogenic NORSE should be reported separately from NORSE of ultimately identified etiology, but the former should be distinguished from cases with incomplete diagnostic evaluation. Likewise, a multidisciplinary care model is imperative to achieve the best patient care. In addition, the eight LTO domains proposed could be used in other forms of RSE.

AUTHOR CONTRIBUTIONS

Poul H. Espino: Conceptualization; data curation; investigation; writing—original draft (lead). Krista Eschbach: Conceptualization; data curation; project administration (lead); methodology (supporting); writing—review and editing. Leah J. Blank, Mackenzie C. Cervenka, Eyal Muscal, Raquel Farias‐Moeller, Emily J. Gilmore, Margaret T. Gopaul, Hiba A. Haider, Aurelie Hanin, Lawrence J. Hirsch, Marissa A. Kellogg, Gerhard Kluger, Soon‐Tae Lee, Alexandria E. Melendez‐Zaidi, Vincent Navarro, Audrey C. Oliger, Elena Pasini, Gitta Reuner, Cynthia M. Sharpe, Zubeda B. Sheikh, Leon Steigleder, Claude Steriade, Coral M. Stredny, Adam Strzelczyk, Olga Taraschenko, Andreas van Baalen, Sarah A. Vinette, Ronny Wickström, Nora W. Wong, and Jiyeoun Yoo: Conceptualization; methodology (supporting); writing—review and editing. Teneille E. Gofton: Conceptualization; data curation; methodology (lead); project administration; supervision; writing—original draft preparation; writing—review and editing (lead).

CONFLICT OF INTEREST STATEMENT

T.E.G. has received support from the Robert N. Kohn NORSE Family Registry Memorial Research Fund and from the Canadian Institutes of Health Research, Lawson Internal Research Fund, and Academic Medical Organization of Southwestern Ontario for unrelated research. M.C.C. has received support from the Johns Hopkins Center for Refractory Status Epilepticus and Neuroinflammation; has served as a paid consultant for Nutricia/Danone, Vitaflo/Nestle Health Science; and has received honoraria from Nutricia/Danone, Vitaflo/Nestle Health Science and royalties from Demos Health/Springer Publishing Company during the period of the research activity and generation of the current report. A.H. has received support from the Paratonnerre Association, the Swebilius Foundation, the Servier Institute, the Philippe Foundation, and the NORSE Institute for NORSE‐Related Research. H.A.H. has received support from the NORSE Institute Medical as an advisory board member, Springer Publishing, and UpToDate (research funding: NINDS R21NS137117 and NCATS UG3TR004501‐01A1). S.‐T.L. has received support from Roche/Genentech (steering committee), Advanced Neural Technologies (advisory board), GC Pharma, and Celltrion, outside the submitted work. A.S. has received support from Angelini Pharma, Biocodex, Desitin Arzneimittel, Eisai, Jazz Pharmaceuticals, Takeda, UCB Pharma, and UNEEG medical. C.M.Sh. has received support from the Pediatric Epilepsy Research Foundation. C.S. has received support from the Epilepsy Study Consortium for work provided to Alterity, Baergic, Biogen, BioXcel, Cerebral, Cerevel, Eliem, Equilibre, Janssen, Longboard, Lundbeck, Marinus, NeuCyte, Neurocrine, Neuroelectrics, Ono, SK Life Science, and UCB; grants from NINDS, AES, UCB, and Parekh Centre for Interdisciplinary Neurology outside the submitted work; speakers bureau fees on behalf of SK Life Sciences and Neurelis; and fees for editorial work for Dynamed. The remaining authors have no conflict of interest. No funding was received to directly support this article. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Espino PH, Eschbach K, Blank LJ, Cervenka MC, Muscal E, Farias‐Moeller R, et al. New onset refractory status epilepticus: Long‐term outcomes beyond seizures. Epilepsia. 2025;66:988–1005. 10.1111/epi.18267

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.