The importance of acknowledging diagnostic uncertainty in patients with new-onset paroxysmal spells

The aims of this study were to evaluate the frequency of paroxysmal spells of indeterminate nature (PSIN) in a large cohort of children and adults with suspected new-onset seizures, to evaluate the reasons for including patients in this category, and to calculate the rate of erroneous diagnoses if the epileptologists were compelled to label those events as epileptic seizures or nonepileptic paroxysmal spells.

Patients identified for this study participated in a prospective study evaluating patients with suspected new-onset unprovoked seizures. The workup included a detailed history and a thorough description of the spells, a 3-hour video EEG recording, and an epilepsy protocol brain MRI. Based exclusively on a detailed description of the ictal events, two epileptologists were asked to independently classify each patient into those with a definite diagnosis of unprovoked seizures or a definite diagnosis of a nonepileptic paroxysmal spells (group 1) and those with PSIN (group 2).

A total of 1880 consecutive patients were enrolled with 255 (13.6%) included in the PSIN group. Patients with PSIN were significantly younger than those with a definite diagnosis, and PSIN were significantly more frequent in children with developmental delay. The most common reason for including patients in the PSIN group was the inability to categorically discriminate between a seizure and a nonepileptic mimicker. When the raters were compelled to classify the spells in the PSIN group, the frequencies of erroneous diagnoses ranged between 32% and 38%. The final diagnoses on those patients were made based on the results of the EEG, MRI, and follow-up visits.

Our data indicate that a diagnostic category of PSIN should be recognized and ought to be used in clinical practice. Acknowledging this uncertainty will result in lower frequencies of erroneous diagnoses, possible stigma, and potential exposure to unnecessary antiseizure medications.