This website is intended for healthcare professionals

Paroxysmal non-epileptic seizures in children: recognition and approach to diagnosis

Posted in Paedatric Neurology on 11th Dec 2012


Dr Tekki S Rao

Dr Tekki S Rao DCH, DNB(paeds), MRCPCH is a Consultant in Paediatrics at Luton and Dunstable University Hospitals. He graduated from Kurnool Medical College, India and was trained further in paediatrics acquiring additional qualifications, DCH from AP University of Health Sciences and DNB (Paediatrics) from National Board of Examinations, New Delhi. He passed MRCPCH in 2002 and obtained specialist training in Paediatrics from Leicester University Hospitals. He has a special interest in Paediatric Epilepsy and Paediatric Neurology. He has honorary attachment with the Paediatric Neurology Department, Addenbrookes Hospital, Cambridge, UK.

Correspondence to:
Dr Tekki S Rao , Department of Paediatrics Luton and Dunstable University Hospitals Foundation NHS Trust Luton, Bedfordshire, LU4 0DZ, UK. Email: Tel: +44 (0)1582 497203.
Acknowledgements: Special thanks to Dr Anna Maw, Consultant Paediatric Neurologist at Department of Paediatric Neurology, Addenbrookes Hospital, Cambridge, who has helped in editing this article.

Non-epileptic seizures (NES) constitute an important differential diagnosis for epileptic seizures in all age groups. NES should be carefully considered and should be ruled out before making a diagnosis of epilepsy. Confident diagnosis of NES in children is often more challenging than making a positive diagnosis of epilepsy. A significant proportion of children suspected of epilepsy or even those who have been labelled with a definite diagnosis of epilepsy or even refractory epilepsy, have never had an epileptic seizure.1


The prevalence of epilepsy is estimated to be 4-5/1000 children in the European and North American population. Up to 30% of these individuals may have a misdiagnosis. The rate of misdiagnosis in epilepsy in adults is estimated to be 25% in one study.2 There is no data to provide an estimate of rates of misdiagnosis in children but at an enquiry of a tertiary paediatric neurology service in the UK, the misdiagnosis rate was found to be 32%. A significant proportion of misdiagnoses comprise NES which are mislabelled as epileptic seizures. Syncope is more prevalent than either epilepsy or psychogenic seizures and is common across all age groups. Prevalence of dissociative (psychogenic) seizures in adults is estimated to be between 2 and 33 per 100,000 population.11 Psychogenic seizures are more prevalent in females (75%) and typically begin in late teens. Common causes of misdiagnosis are; poor history taking, diversity of presentation of epileptic events, no sensitive or specific diagnostic tests available for epilepsy and many imitators that are confused with the diagnosis of epilepsy.

How important is it to get it right?

The cost of misdiagnosing NES as epilepsy falls mainly on the National Health Service in the UK, but also on education, social care and wider society. Beyond economic factors one should also be aware of potentially irreparable damage caused as a result of making the wrong diagnosis. This diagnosis makes an immeasurable negative impact on the psychological wellbeing of the child and family, loss of school days, loss of parent’s working days, impact of drugs on an individual’s cognition and the long-term impact on career, driving and lifestyle etc.2 In addition to this, there is a risk of missing another serious diagnosis such as Long QT syndrome (Figure 1) that leads to death from ventricular tachyarrhythmias (torsades de pointes).3 Therefore, it is the responsibility of the physician to identify and separate NES from true epilepsy right at the beginning. Hence, the National Institute for Health and Clinical Excellence (NICE) recommends that all children and young adults who have been suspected of epilepsy should be seen by an expert in epilepsy.4

Figure 1. ECG in patient with corrected prolonged QT interval. QTc of more than 0.44 sec is significant.

Diversity of non-epileptic events

There are a wide number of conditions which can mimic epilepsy and these have been extensively described in text books and review articles.1,5,6,7 These conditions can be grouped by the system that is probably involved (Table 1) and by symptom of presentation (Table 2).

Challenges in differentiating NES from epilepsies

Every condition in the NES group mimics an epileptic condition. There is no single symptom which is generally linked to the central nervous system that can confidently be excluded as epileptic. Even certain symptoms which are only distantly related to the CNS (isolated vomiting, hiccups, sweating, facial flushing and a feeling of unfamiliarity) could also be manifestations of epilepsy.

Particular challenges include;

  • Symptoms are often of very short duration and difficult to capture on a video.
  • The first account witnesses, who are often very frightened parents, may give a poor description of the event. It is quite challenging even for professionals trained in epilepsy to give a clear description of a paroxysmal event.
  • Infants and children younger than four or five years are usually unable to provide a useful subjective description of their own symptoms.
  • There is no reliable diagnostic test to differentiate NES from epilepsies. The EEG is a very poorly sensitive and specific test in diagnosing or ruling out epilepsy.
  • Paediatricians are often pressurised by carers, school teachers, paramedics and other allied professionals who observe and report seizures.

Syncopes are the most common non-epileptic disorders misdiagnosed as epilepsy.2 Conversion disorder was seen in children over five years of age, becoming the most common type of paroxysmal non-epileptic event among adolescents.7 Common conditions that paediatricians encounter regularly in their clinical practice that pose diagnostic challenge are shown in Table 3.

Clues to the diagnosis

Table 4 outlines the key features of more common NES conditions. Features such as incontinence, tongue biting and external injury do not help in distinguishing psychogenic seizures from epilepsy. Seizure duration of more than two minutes, closed eyes, thrashing, pelvic thrust, opisthotonus, fluctuating course and recall for a period of unresponsiveness suggest psychogenic seizures although may also occur with less frequency in epilepsy. Ictal observation gives useful clues to the diagnosis (Table 5).

Visual Symptoms: Aura in migraines can be visual, sensory or motor and may suggest epilepsy. Migrainous aura and epilepsy can be distinguished when the child can describe symptoms or draw pictures of their visual aura. It is good practice in clinic to encourage children to draw what they have visualised. Visual migrainous auras are monochromatic, angulated, bright and scintillating. They start in the centre and spread to the periphery and leave scotomata. The duration of aura may last up to one hour. In contrast, focal onset seizures of occipital lobe have visual manifestations that are described as circular, amorphous, multicoloured and the duration is seconds up to a maximum of two to three minutes. They appear in the periphery of the visual field.


Investigations should be individually tailored and carefully selected. Investigations should provide supplementary evidence to support the clinical diagnosis but are rarely diagnostic in themselves. False positive results may bias a diagnosis.

Ictal video: Video recording by parents on their mobile phone or on home video equipment assisted by community nurses or in-patient video telemetry can be an invaluable investigation in clinching the diagnosis of NES. Nowadays, most people have ready access to video technology to enable them to take a short high definition recording. In some conditions such as sleep events a longer duration of recording may be required. In our experience admitting patients for video monitoring of paroxysmal events has been a fruitful approach.9

ECG: In children and young adults a 12 lead ECG should be considered in cases of diagnostic uncertainty4 and should be undertaken in all children with suspected syncopes. More extensive cardiac investigations such as echocardiogram, prolonged ECG recording (up to seven days), cardiac memo and tilt table testing may be indicated in individual cases of NES suspected to be of cardiovascular origin.

EEG: Since the EEG is a poorly sensitive and specific investigation in the diagnosis of epilepsy, it should be used with great caution in NES. EEG should not be performed in cases of probable syncope because of the possibility of false positive results.4 The National Institute of Clinical Excellence says that an EEG should not be used to exclude the diagnosis of epilepsy in a child, young person or adult in whom clinical presentation supports a diagnosis of non-epileptic event.4 An approach to positive diagnosis of non-epileptic seizure should not come from ruling out epilepsy by obtaining negative EEG. Ictal EEG with simultaneous video monitoring is an extremely useful investigation in psychogenic seizures. Video telemetry, if available, is more diagnostic in psychogenic seizures. Telemetry facilities are limited in the UK and not available at all secondary and many other tertiary care paediatric neurology services. Ambulatory EEG can be useful if a paroxysmal event can be captured within the time frame.

Neuro-imaging: This is of limited help in establishing a diagnosis but could be indicated in suspected neurological conditions such as Arnold-Chiari malformation, suspected raised intracranial pressure and intracranial space occupying lesions.

Sleep studies: Sleep studies are indicated in obstructive sleep apnoea, narcolepsy and REM/non-REM sleep disorders. These could be supplemented by video recording when nocturnal epilepsies e.g. Autosomal Dominant Nocturnal Frontal Lobe Epilepsies (ADNFLE) can be identified and distinguished from NES.

Genetics: Molecular genetic tests are gaining importance in establishing diagnosis of NES especially channelopathies (Long QT syndrome, BPTI, BPVC, Episodic ataxias, Paroxysmal tonic up gaze and Hemiplegic migraine). Calcium, sodium, potassium channel genes will be supportive in strongly suspected cases.

Other investigations: Other useful investigations are pH or impedance studies in gastro-oesophageal reflux and Sandifer syndrome, blood tests such as calcium, electrolytes, magnesium and blood sugar, when indicated.

Unhelpful investigations: Serum prolactin and creatine kinase are not useful investigations in differentiating epilepsy from psychogenic seizures. NICE Guidelines do not recommend measuring prolactin in the diagnosis of epilepsy.4


Diagnosing NES and differentiating them from epilepsy is almost always based on clinical history. It often requires lengthy discussions and interviews of parents, patients and witnesses.6 During the consultation it may be necessary for the physician to imitate and demonstrate physically some of the paroxysmal events to get a clearer picture of the condition. Some experienced authors promote the practice of showing video recordings of different epileptic and non-epileptic seizure examples to the parents to discover which, if any, resemble their own child’s attacks – the ‘that’s it’ phenomenon.10

It is a good clinical practice to encourage parents and carers to obtain video records of these episodes where possible when there is a clinical suspicion of the nature of seizures. This should remain the first line of investigation in confirming the diagnosis. Video recording can be reviewed repeatedly by the physician and a peer review and opinion from experts can be obtained if the diagnosis is still unclear.

ECG should be obtained in all convulsive seizures and in cases where cardiac cause is suspected, referral to cardiologist and cardiovascular investigations should be arranged. All other Investigations (EEG, neuro-imaging, and blood investigations) can only support or refute a clinicians’ suspicion and it is advised to use these investigations very judiciously.

Wait and watch policy pays rewards in diagnosis and management of NES. Time and patience is a more valuable investment in making the diagnosis. Time spent in gathering more information from all sources (e.g. School teachers, peer students, paramedics), obtaining video records, seeking peer review and expert opinion is rewarding and worthy. Haste in making a diagnosis should be avoided because it is often very difficult for everyone concerned to withdraw a diagnosis of epilepsy once the label is given.

Every paroxysmal condition should be analysed with suspicion. When in doubt, even a few years down the line, clinicians should have no hesitation to revisit the diagnosis and seek peer review.

Treatment will depend on the nature of the underlying condition, however recognition of NES and reassurance early in the course of presentation would facilitate more appropriate management. Most of these conditions do not require management with medicines, but some of them would require psychotherapy in particular, psychogenic seizures. Recognition and explanation of diagnosis are important components of management of these conditions.


There are large numbers of neurological, cardiac, psychogenic and other miscellaneous disorders that result in paroxysmal clinical events. Varied but similar presentations akin to epilepsy lead to misdiagnosis. Clinicians should be aware of differentiating features and take a thorough history. Video recording should supplement the history when the clinical picture is not clear and along with the judicious use of investigations. Nevertheless, misdiagnosis is common and may have profound psychological, physiological and socio-economic consequences to the patient, parents and economic burdens to health services.


  1. Stephenson J, Whitehouse WP and Zuberi S. Paroxysmal non-epileptic disorders: differential diagnosis of epilepsy. In: Sheila J Wallace (ed.) Epilepsy in Children. 2nd Edition. Arnold, London. 2004.
  2. Smith D, Defalla BA, Chadwick DW. The misdiagnosis of epilepsy and management of refractory epilepsy in a specialist clinic. Q J Med 1999;1:15-23.
  3. Hobbs JB, Peterson DR, Moss AJ, et al. Risk of aborted cardiac arrest or sudden cardiac death during adolescence in the long-QT syndrome. JAMA. Sep 13 2006;296(10):1249-54.
  4. The Epilepsies. The diagnosis and management of the epilepsies in adults and children in primary and secondary care. Clinical Guidelines, CG137. NICE guidelines for epilepsy. January 2012.
  5. 5. Rugg-gun FJ. Non-epileptic paroxysmal neurological and cardiac events: the differential diagnosis of epilepsy. In: Epilepsy 2009; From Bedside to Bedside; A practical guide to epilepsy. Lecture notes 12th epilepsy teaching week 18-20 September 2009. Oxford; 165-180.
  6. Differential diagnosis of paroxysmal events: epilepsy and nonepileptic seizures. In: Panayiotopolous CP(Ed.). The Epilepsies. Seizures, syndromes and management.2nd edition. Blandon Medical Publishing. 2005;1-7.
  7. Kotagal P, Costa M, Wyllie E, Wolgamuth B. Paroxysmal nonepileptic events in children and adolescents. Paediatrics 2002;110:E4-6.
  8. Lamport T, von Brevern M. The eye movements of syncope. Neurology 1996;46:1086-8.
  9. Rao TS, Stevens E. In-patient video monitoring can be an invaluable diagnostic tool of epilepsy in the hands of a hospital paediatrician. DMCN. January 2012;54:55-89.
  10. Stephenson JB. Fits and Faints. MacKeith Press, London. 1990.
  11. Benbadis SR, Allen HW. An estimate of prevalence of psychogenic non-epileptic seizures. Seizure 2000;9:280-1.
Download this Article