Statistics from Altmetric.com
Do et al present a patient with paroxysmal sympathetic storm, and include a sample of the patient's ictal EEG recording.1 It is stated that the absence of clear epileptiform activity serves as evidence against these episodes being epileptic in nature. However, if indeed the EEG recordings during an attack show significant slowing, as depicted in the EEG sample, this EEG change from a presumably normal EEG background rhythm at other times, would rather indicate that these events are, indeed, of seizure origin.
Deep seated epileptic foci very often do not project any sharply configured waveforms to the scalp surface; neither do they provide reliable localising information and can appear rather in a generalised fashion. In such cases, any reproducible and reliably observable EEG changes from the background rhythm that coincide with the clinical event are typically interpreted as evidence in favour of epileptic activity. Possible candidates for surgical treatment of their epilepsy would be further investigated with implanted electrodes in an effort to obtain EEG recordings from the closer vicinity of the presumed focus. This is obviously not a justifiable approach in this patient as surgical resection is not an option in this location.
Neither should successful treatment of specific symptoms during an attack with medication other that antiepileptic drugs necessarily be interpreted as evidence against the event being epileptic. In this case report, the patient's autonomic disturbances during the episode did respond to labetalol; however, this does not necessarily exclude a possibly epileptic origin as labetalol could only have obscured the clinically observable manifestations.
Bromfield et al reply:
We appreciate Bernath's thoughtful comments concerning our case report of using labetolol to treat paroxysmal sympathetic storm. We agree that ictal EEG patterns can be quite variable and subtle, and that paroxysmal slowing can correspond to a seizure even without observable sharp waves or spikes. In this context, rhythmicity is generally viewed as a major characteristic of an ictal EEG, usually with a typical frequency evolution, sometimes accelerating after onset and usually slowing before stopping.1-1 These characteristics were not seen in our patient. Furthermore, we should clarify that the background between these episodes was not normal, but rather showed a 6–8 Hz, somewhat disorganised posterior rhythm, with intermixed diffuse theta and some low voltage delta. We therefore interpreted the higher amplitude slowing during attacks, as seen in the previously published figure, as an arousal response. We have enclosed with this communication a sample from the same EEG study that shows the typical background activity followed by a definite arousal, corresponding to the notation of “noise”; this arousal is similar to that recorded during attacks (fig 1-1). In addition, we should have made it clear that, although later alert and responsive both during and between attacks, at the time this EEG was performed, before completing treatment of the shunt infection, he was moderately lethargic between attacks, and became more alert and agitated during them. Space limitations prevented our inclusion of this “baseline” EEG pattern during the initial report.
The fact that the patient responded to autonomic agents rather than to antiepileptic drugs is not a definitive argument against an epileptic origin, as noted by Bernath, although it provides at least circumstantial evidence. Furthermore, it seems unlikely that a diffuse EEG change, as shown in the original figure, would have only autonomic manifestations, and would be completely suppressed clinically by labetolol. Simple partial seizures with only autonomic manifestations would be more likely to show a unilateral temporal discharge or an unchanged EEG.1-2
For these reasons, we think that our patient in fact had paroxysmal sympathetic storm rather than multiple daily, prolonged autonomic seizures. As noted by Bernath, answering this question definitively would have required intracranial electrode placement, which was not clinically indicated in this case.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.