The importance of acknowledging clinical uncertainty in the diagnosis of epilepsy and non-epileptic events

Abstract

Background: Failure to recognise diagnostic uncertainty between the epilepsies and non-epileptic events may be a factor in high rates of misdiagnosis. Aims: To explore the results of acknowledging diagnostic uncertainty in a cohort of children presenting with paroxysmal events. Methods: Children (29 days-16th birthday) with new presentations of paroxysmal disorders were ascertained through outpatients, admissions, and accident and emergency over a two year period in a district hospital with a catchment population of 500 000. Cases were classified by diagnosis at entry and 6-30 months later. A random selection of cases was independently assessed. Results: A total of 684 cases were ascertained. Attacks were initially classified as febrile seizures (n = 212), acute symptomatic epileptic seizures (n = 5), epilepsies (n = 83), unclassified (possible epilepsy) (n = 90), isolated epileptic seizures (n = 51), and non-epileptic events (n = 243). Case review enabled reclassification of 61 of those initially unclassified-31 to an epilepsy and 27 to non-epileptic events. In 29 the final diagnosis was never clarified. These were 23 cases with confusing or absent histories and six with short lived seizure clusters. Prognosis for these 29 cases was good; 75% had been discharged. None were on long term medication. The diagnosis in the 131 cases confirmed as epilepsy was stable. Independent review of a random sample showed full concordance with one neurologist and 20% uncertainty with another. Conclusion: In addition to definite epilepsy or non-epileptic events it is helpful to recognise a group of cases where the diagnosis is uncertain-unclassified paroxysmal events. Reassessment of these cases enables accurate diagnosis and may prevent a hasty and incorrect diagnosis of epilepsy.

Misdiagnosis in epilepsy: a review and recognition of diagnostic uncertainty

“…Where present, uncertainty should be shared with the patient and other physicians involved in their care. Beach and Reading [7] recognized this and proposed the term ‘unclassified paroxysmal event’. We also propose that where a decision is made to manage the patient on the basis of ‘probable’ but not definite diagnosis of epilepsy, this uncertainty should be documented and the diagnosis reviewed at regular intervals.…”

“…In a subsequent British prospective study by Beach and Reading [7], new presentations of paroxysmal disorders were reviewed 6-30 months after the initial diagnosis by the same physician. The events were initially classified as an isolated seizure (51), epilepsy (83), possible epilepsy (90) or non-epileptic events (243).…”

Paediatric new‐onset seizure clinic in Australia: Experience and lessons learnt

“…A paediatric neurologist review, with integration of the clinical profile and information from EEG, results in a high level of accuracy of diagnosis in an NOSC: none of the children in our cohort had their initial EP+ or EP− diagnosis revised at follow‐up. Beach et al . report the same accuracy.…”

Guidelines, training, audit, and quality standards in children's epilepsy services: Closing the loop

“…26 There may be an increasing acceptability of uncertainty as a diagnostic state, with clinicians more inclined towards the active management of uncertainty rather than perhaps historic approaches where a diagnosis of an epilepsy was assumed, with subsequent trials of anti-epileptic treatment. 27 The effect of this may be more accurate diagnosis as compared to even a decade ago.…”