Abstracts

RATIONALE: Avoiding an inappropriate diagnosis of epilepsy is of critical importance, and it is well accepted among epileptologists that more damage is done by overdiagnosis than by underdiagnosis of epilepsy. Nevertheless, epileptologists frequently encounter EEGs that have been [dsquote]over-read[dsquote] and result in erroneous diagnoses of epilepsy. We collected a series of such records in order to analyze what patterns tend to be over interpreted.
METHODS: Cases were collected over a 2 year period. We collected patients who were diagnosed with nonepileptic events but carried a diagnosis of epilepsy. We selected those who had had an [dsquote]abnormal[dsquote] EEG, and then made every effort to obtain both EEG reports and actual tracings. Tracings were then reviewed by Board-certified (ABCN) electroencephalographers.
RESULTS: Fifteen EEGs were collected, which were clearly over-read as containing epileptiform abnormalities. Well described benign (epileptiform) variants were found in 3 cases: wicket spikes (1), hypnagogic hypersynchrony (1), and hyperventilation-induced slowing (1). There were no tracings with small sharp spikes, psychomotor variants, or fourteen & six positive spikes. In the other 12 records, the over-read patterns were simple fluctuations of sharply contoured background rhythms (abundant examples will be shown).
CONCLUSIONS: We could only obtain a small fraction of the requested EEGs read as abnormal, so an estimate of frequency is not possible. The problem of EEG over-interpretation is almost certainly under-reported, and contributes to erroneous diagnoses of epilepsy. More so than well described normal variants that can look epileptiform, the most common misinterpreted patterns are common variations of normal rhythms.