Functional seizures: clues not to confuse with epilepsy

A 27-year-old woman arrives at the emergency department after an episode of generalized movements that lasted several minutes. The pre-hospital team has already given medication for the seizure, and on arrival she continues to have shaking that comes and goes, with her eyes tightly closed. She has had a diagnosis of epilepsy for years, takes two anticonvulsants and has already been admitted to intensive care for "status epilepticus". The attending physician considers escalating treatment.

This scene may represent difficult-to-control epilepsy. It may also represent a functional seizure, called in the literature psychogenic nonepileptic seizure. The two conditions can coexist in the same person, and an error in either direction has consequences: treating a functional seizure as status epilepticus exposes the patient to medications and procedures that do not help, and labeling epilepsy as functional leaves real seizures untreated.

This text gathers what studies from several countries show about the frequency of functional seizures, the signs that help at the bedside, the role of video-EEG, the risk of confusing them with status epilepticus and how to communicate the diagnosis.

Key points

• Functional seizures are common in epilepsy centers and often take years to be correctly diagnosed.
• A single sign rarely confirms the diagnosis, but some semiological findings greatly increase the probability.
• Video-EEG capturing the typical event is the reference standard.
• Prolonged functional seizures are mistaken for status epilepticus and treated with medications that carry risks.
• The way the diagnosis is communicated influences acceptance and outcome.

What functional seizures are

Functional seizures are episodes of altered behavior, consciousness, motor activity or sensation that resemble epileptic seizures but occur without the corresponding epileptic electrical activity. An updated review describes these events as paroxysmal episodes with motor, sensory, mental or autonomic manifestations, diagnosed after medical causes such as epilepsy, syncope and stroke are ruled out (Lanzillotti and colleagues, 2021).

The name has changed over time. Pseudoseizure, hysterical seizure and psychogenic seizure are old terms, still used in articles, but they carry judgment. The expression "functional seizure" describes a problem in the functioning of the nervous system and avoids suggesting that the person is faking or controlling the symptoms. A pediatric review describes the seizures as observable paroxysmal changes in behavior or consciousness, similar to epileptic seizures, without concomitant electroencephalographic changes (Operto and colleagues, 2019).

The main points of the concept are:

• the episodes are involuntary, and the person usually suffers from them;
• the absence of epileptic discharge is confirmed, preferably by video-EEG;
• the diagnosis relies on positive signs, not only on ruling out epilepsy;
• functional and epileptic seizures can coexist in the same person;
• psychiatric comorbidities and a history of trauma are frequent, but not always present.

The last point deserves care. A study on communication with families of young people advises saying that, in up to a third of cases, no stressor or mental health condition is identified (Patel and colleagues, 2024). The absence of a known trauma, therefore, is far from ruling out the diagnosis.

Why the error is so frequent

Functional seizures are common in specialized services. The updated review points out that they affect between 20 and 30 out of every 100 patients seen at epilepsy centers (Lanzillotti et al., 2021). A review on prognosis cites an estimated incidence of 1.4 per 100,000 people per year and notes that about 20 out of every 100 patients referred to tertiary centers for refractory seizures have this condition (Durrant et al., 2011).

The path to diagnosis is often long. The task force of the International League Against Epilepsy reports that most patients are first diagnosed with epilepsy and that the average delay between seizure onset and diagnosis is around 5 years (Kanemoto et al., 2017). A review on video-EEG cites an average interval of 7 to 10 years until a definitive diagnosis and concludes that the index of suspicion is low and misdiagnosis is common (Gedzelman and LaRoche, 2014).

Several factors explain this delay:

• the physician rarely sees the seizure, and the description given by witnesses is imprecise;
• the epilepsy label, once recorded, tends to be repeated without review;
• a routine EEG can be normal in someone with epilepsy and nonspecifically abnormal in someone with functional seizures;
• access to video-EEG is limited in many places;
• the functional diagnosis is still seen by some professionals as a diagnosis of exclusion.

This scenario is a classic example of diagnostic overshadowing: the old label comes to explain any new episode. The text on stigma and language discusses how labels shape management, and the text on overdiagnosis helps to think about the weight of a diagnosis that stays.

Routine EEG misleads in both directions

The review on video-EEG notes that 30 out of every 100 people with epilepsy have a normal EEG at presentation and that benign variants or nonspecific changes appear in more than 10 out of every 100 healthy individuals (Gedzelman and LaRoche, 2014). The authors also cite a retrospective study in which about 0.5 out of every 100 healthy people, without epilepsy or functional seizures, had unequivocal epileptiform discharges (Gedzelman and LaRoche, 2014). An abnormal EEG, therefore, is far from proving epilepsy, and a normal EEG is far from ruling it out.

Semiological clues at the bedside

Observing the seizure is the richest source of information, as long as it is done methodically. An Australian study reviewed 742 events recorded in 140 video-EEG studies in adults, carried out from May 2009 to August 2016 (Duncan et al., 2022). A fluctuating course had an odds ratio of 37.37 for functional seizures, head shaking had an odds ratio of 2.95 and pelvic thrusting had an odds ratio of 4.28 (Duncan et al., 2022). A fluctuating course had the highest sensitivity, 76.16 out of every 100, and back arching, with an odds ratio of 1.06, showed no significant association with functional seizures (Duncan et al., 2022).

The review on video-EEG lists other findings associated with functional seizures: eye closure, especially prolonged, eye movements that follow gravity, stuttering during the seizure and whispered speech after the event (Gedzelman and LaRoche, 2014). The updated review notes that a longer duration of episodes is a common finding in functional seizures, but that no sign is pathognomonic (Lanzillotti et al., 2021).

The table below summarizes the clues described in the references:

Finding	What it suggests	Caution
Fluctuating course, coming and going	Functional seizure	It is the sign with the highest sensitivity in the Australian study
Eyes tightly closed during the event	Functional seizure	Eye closure can also occur in other situations
Head shaking and pelvic movement	Functional seizure	Frontal lobe seizures can have similar movements
Back arching	Poorly discriminative	It showed no significant association in the Australian study
Tongue biting and falls with injury	Traditionally epilepsy	Also occurs in functional seizures
Whispered speech after the event	Functional seizure	It needs to be interpreted together with the rest of the picture

Classic signs that mislead

Some signs that common sense associates with epilepsy also appear in functional seizures. The updated review reports that between 8 and 31 out of 100 patients with functional seizures suffered injuries during the seizures, such as tongue biting, bumps and falls (Lanzillotti and colleagues, 2021). Loss of sphincter control and seizures that appear to arise during sleep were also described in these people (Lanzillotti and colleagues, 2021). Reasoning that relies on a single "typical" sign carries risk in both directions.

Simple tests during the event

A good practice proposal for video-EEG suggests testing consciousness during the event, for example by asking the person to show their tongue, and testing memory afterwards by asking them to recall a word said during the seizure (Whitehead and colleagues, 2017). The same document suggests considering bedside avoidance tests, such as resistance to eye opening (Whitehead and colleagues, 2017). These tests are observations, not traps: the goal is to describe accurately what happens, with respect for the person.

Video-EEG: the reference standard

Video-EEG lets you see the seizure and the tracing at the same time, and it is described as the reference standard for diagnosing functional seizures (Gedzelman and LaRoche, 2014). The pediatric review also highlights video-EEG and home video as useful tools, especially in children who have both epileptic and functional seizures (Operto and colleagues, 2019).

The good practice proposal recommends obtaining consent, starting the recording early, collecting a detailed description of the semiology with the patient and witnesses and, when capturing an event, asking whether it is typical of the usual episodes (Whitehead and colleagues, 2017). The authors point out as one of the greatest dangers the overinterpretation of benign EEG phenomena or movement artifacts as epileptiform activity (Whitehead and colleagues, 2017).

Video-EEG also corrects diagnoses in the opposite direction. In a tertiary service in the United Kingdom, among 444 video-telemetry reports with usual events recorded from 2019 to 2022, 4.7 out of 100 patients referred as functional seizure received a diagnosis of epilepsy or another diagnosis (Dudley and colleagues, 2024). The authors cite a revision rate of 3.3 out of 100 at another center and describe the most confused groups, such as frontal lobe seizures and insular or temporal seizures with marked autonomic symptoms (Dudley and colleagues, 2024).

Complementary tests have a limited role

The review on video-EEG reports that a rise in serum prolactin within 30 minutes of event onset can help differentiate generalized tonic-clonic seizure from functional seizure, but only as a complementary test (Gedzelman and LaRoche, 2014). Imaging and laboratory tests, on their own, rarely clarify the diagnosis. The reasoning of pretest probability helps explain why a test with moderate performance changes the decision little when the initial hypothesis is strong.

When epilepsy and functional seizure coexist

Some patients have both types of event. This combination is one of the most difficult situations, because confirmation of one type of seizure tends to erase the possibility of the other. People with confirmed epilepsy can develop functional seizures over time, and people with a functional diagnosis can also have epileptic seizures that only appear when medication is withdrawn, as the British study on reclassifications describes (Dudley and colleagues, 2024).

The safest strategy is to characterize each type of event separately. Asking the family to describe, or film, each seizure pattern, and recording in the chart which episodes were confirmed by video-EEG and of which type, helps prevent every new event from receiving the same label. The good practice proposal suggests precisely starting the report with the updated history of all types of seizure (Whitehead and colleagues, 2017).

Cognitive traps

Reasoning during a crisis is vulnerable to several biases. Anchoring on the old epilepsy diagnosis leads the team to escalate medications without revisiting the hypothesis. The opposite bias appears when a previous functional diagnosis leads the team to dismiss a new event with different features. On-call fatigue and frustration with repeated visits favor shortcuts in both directions.

Some practices help reduce these risks. Describing the seizure with concrete words, instead of using only the label, preserves useful information for whoever comes next. Asking whether the current episode resembles the previous ones, and how it differs, forces a comparison. Discussing the visits that ended in reclassification, without a tone of blame, builds the team's collective memory, as the morbidity and mortality sessions show.

The risk of treating it as status epilepticus

The error with the most immediate consequence is treating a prolonged functional seizure as status epilepticus. An individual data analysis of two large trials on status epilepticus, RAMPART and ESETT, included 980 patients aged 8 or older treated for status epilepticus; 79 of them, or 8.1 per 100, were discharged with a final diagnosis of prolonged functional seizure (Jungilligens and colleagues, 2021). The proportion reached 20.1 per 100 among adolescents and young adults (Jungilligens and colleagues, 2021).

The same study describes the harms. Adverse effects, including respiratory depression and intubation, were recorded in 26 per 100 patients with functional seizure who received benzodiazepines in RAMPART and in 33 per 100 of those who received second-line medication in ESETT (Jungilligens and colleagues, 2021). In both trials, one in five patients with prolonged functional seizure was admitted to intensive care (Jungilligens and colleagues, 2021).

The authors suggest a cumulative momentum effect of treatment: when the first medication is given outside the hospital, the next team tends to keep escalating (Jungilligens and colleagues, 2021). This echoes the scene of the 27-year-old patient, who arrives already medicated and with a history of intensive care.

Questions before escalating

Faced with a prolonged seizure, some questions help curb the automatic response without failing to treat those who need it:

1. Is the course continuous or fluctuating, with pauses and resumptions?
2. Are the eyes tightly closed or open?
3. Are there signs of hypoxemia, cyanosis or autonomic changes consistent with a generalized seizure?
4. Was the pattern of previous episodes recorded on video-EEG?
5. Did the response to previous medications make sense for epilepsy?

These questions fit into a short diagnostic pause and should be discussed with someone experienced in epilepsy. The decision to treat status epilepticus remains a priority when doubt persists, because true status epilepticus is an emergency.

Communicating the diagnosis and what comes next

Communicating the diagnosis is part of treatment. In a qualitative study with 13 young people aged 12 to 18 and 13 parents, participants reported intense fear about symptoms, diagnosis and prognosis, valued reassurance, listening and time for questions, and criticized medical jargon, brief conversations, minimization of symptoms and lack of written material (Patel and colleagues, 2024).

In adults, the way of communicating also seems to matter. At a Canadian center, among 103 patients with a definitive diagnosis of functional seizure from January 2009 to May 2023, 61 agreed to take part in follow-up (Abi-Nahed and colleagues, 2024). The diagnosis was explained to 97 per 100 by the end of the hospital stay and well accepted by 89 per 100; at contact, 46 per 100 were already seizure-free and 32 per 100 reported that the seizures stopped soon after the diagnosis was communicated (Abi-Nahed and colleagues, 2024). The proportion with daily seizures fell from 38 to 18 per 100 after the diagnosis, over a median follow-up of 51 months (Abi-Nahed and colleagues, 2024).

The prognosis in adults, however, calls for realism. The review on prognosis concludes that fewer than 4 in 10 adults with a recent diagnosis are seizure-free at 5 years (Durrant and colleagues, 2011). This finding reinforces the need for follow-up, psychological treatment and care for comorbidities, discussed in detail in the references.

Elements of a good conversation

Studies on communication point to some recurring elements:

• name the diagnosis clearly, using "functional seizure" or an accepted equivalent term;
• explain that the episodes are real and involuntary;
• show which signs supported the diagnosis, including the video when possible;
• talk about the treatment plan and follow-up, with written material;
• guide family and school on how to act during an episode.

The text on affective biases recalls that the team's frustration with repeated visits can contaminate this conversation. Recognizing that reaction helps keep the focus on the patient.

Children and adolescents

Functional seizures also occur in childhood. The ILAE task force reports that they have been described from age 3, that children lack the female predominance seen in adults, and that seizures with motor inhibition are more common than hypermotor events (Kanemoto and colleagues, 2017). The pediatric review notes that the condition is common and often missed in this age group (Operto and colleagues, 2019).

The prognosis in children tends to be better. In an Australian program with 60 children and adolescents, 45, or 75 out of 100, recovered normal functioning and returned to full-time school, and the global functioning score rose from 41 to 67 (Kozlowska and colleagues, 2018). The review on prognosis estimates remission in about 70 out of 100 children, a more favorable figure than in adults (Durrant and colleagues, 2011).

The explanation given to families makes a difference. In the Australian program, families accepted the diagnosis and engaged in treatment when it was explained how emotional stress, illness and states of high arousal can activate atypical defense responses in the body and brain (Kozlowska and colleagues, 2018).

For young people and families, practical guidance weighs as much as the explanation. In the qualitative study with young people and parents, participants asked for concrete plans for school and guidance on how to respond to a functional seizure in any setting (Patel and colleagues, 2024). A written plan, agreed with the school, reduces unnecessary trips to the emergency department and prevents each episode from being treated as an epileptic emergency. This plan also helps teachers and classmates react calmly, which usually reduces the dramatic attention around the episodes and gives the child back a possible routine.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by looking for the code in ICD-10.

The limits on using the game are described on the NEXO economy page. If you research the teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information under curation and creators, and institutions can learn about the proposal on the partnerships page. Training with differential diagnosis and clues and looking for the right ICD-10 code complete this practice.

Frequently asked questions

What is a functional seizure?

It is an episode that resembles an epileptic seizure but occurs without the corresponding epileptic electrical activity, and it affects between 20 and 30 out of 100 patients seen at epilepsy centers (Lanzillotti and colleagues, 2021).

Which sign most helps raise suspicion of a functional seizure?

The fluctuating course had the highest sensitivity in 742 video-EEG recorded events, from 76.16 out of 100 (Duncan and colleagues, 2022), but no single sign confirms the diagnosis.

Does tongue biting rule out a functional seizure?

That idea is a myth: between 8 and 31 out of 100 patients with functional seizures suffered injuries during the episodes, such as tongue biting, bumps and falls (Lanzillotti and colleagues, 2021).

Can a functional seizure be treated as status epilepticus?

This happens often: among 980 patients treated for status epilepticus in two trials, 79 had prolonged functional seizures (Jungilligens and colleagues, 2021).

Does communicating the diagnosis help the patient?

The data suggest it does: at a Canadian center, 32 out of 100 participants reported that the seizures stopped soon after the diagnosis was communicated (Abi-Nahed and colleagues, 2024).

Take-home messages

• Functional seizures are frequent in epilepsy centers and take years to be diagnosed.
• The diagnosis relies on positive signs, such as a fluctuating course and closed eyes.
• "Typical" epilepsy signs, such as tongue biting, also appear in functional seizures.
• Video-EEG capturing the typical event is the reference standard.
• Prolonged seizures treated as status epilepticus expose the patient to avoidable harm.
• Communicating the diagnosis with clarity and respect is part of the treatment.

References

1. Abi-Nahed R, Li J, Carlier J, et al. Outcome of psychogenic non-epileptic seizures following diagnosis in the epilepsy monitoring unit. Front Neurol. 2024. PMID 38419704.
2. Dudley P, Marquez JP, Farrell F, et al. Functional seizures and their mimics: a retrospective service review of cases from a tertiary video telemetry database. BMJ Neurol Open. 2024. PMID 39119525.
3. Duncan AJ, Peric I, Boston R, et al. Predictive semiology of psychogenic non-epileptic seizures in an epilepsy monitoring unit. J Neurol. 2022. PMID 34550469.
4. Durrant J, Rickards H, Cavanna AE. Prognosis and outcome predictors in psychogenic nonepileptic seizures. Epilepsy Res Treat. 2011. PMID 22937230.
5. Gedzelman ER, LaRoche SM. Long-term video EEG monitoring for diagnosis of psychogenic nonepileptic seizures. Neuropsychiatr Dis Treat. 2014. PMID 25342907.
6. Jungilligens J, Michaelis R, Popkirov S. Misdiagnosis of prolonged psychogenic non-epileptic seizures as status epilepticus: epidemiology and associated risks. J Neurol Neurosurg Psychiatry. 2021. PMID 34362852.
7. Kanemoto K, LaFrance WC, Duncan R, et al. PNES around the world: Where we are now and how we can close the diagnosis and treatment gaps-an ILAE PNES Task Force report. Epilepsia Open. 2017. PMID 29588959.
8. Kozlowska K, Chudleigh C, Cruz C, et al. Psychogenic non-epileptic seizures in children and adolescents: Part II - explanations to families, treatment, and group outcomes. Clin Child Psychol Psychiatry. 2018. PMID 28956479.
9. Lanzillotti AI, Sarudiansky M, Lombardi NR, et al. Updated Review on the Diagnosis and Primary Management of Psychogenic Nonepileptic Seizure Disorders. Neuropsychiatr Dis Treat. 2021. PMID 34113112.
10. Operto FF, Coppola G, Mazza R, et al. Psychogenic nonepileptic seizures in pediatric population: A review. Brain Behav. 2019. PMID 31568694.
11. Patel H, Malave R, Bitting S, et al. Youth and Family Perspectives on Diagnosis Communication About Pediatric Functional Seizures: A Qualitative Study. Pediatr Neurol. 2024. PMID 39353333.
12. Whitehead K, Kane N, Wardrope A, et al. Proposal for best practice in the use of video-EEG when psychogenic non-epileptic seizures are a possible diagnosis. Clin Neurophysiol Pract. 2017. PMID 30214985.