The neurologist and epileptologist at the Mayo Clinic in Jacksonville discussed when to best perform routine EEG in patients with epilepsy.
William Tatum, DO
At the American Epilepsy Society's (AES) annual meeting, held in New Orleans, Louisiana, William Tatum, DO, neurologist and epileptologist at the Mayo Clinic in Jacksonville, gave a presentation on when clinicians should obtain a routine EEG while managing an adult with epilepsy—a topic that provided attendees with practical knowledge to best advance their everyday practice.
Tatum explained that while the question of when is broken down into 3 levels—diagnosis, classifying, and managing—there are some high-yield situations and some low-yield situations that also may be helpful in terms of obtaining a routine EEG.
In the end relative to obtaining in a routine EEG over and over, Tatum concluded that if there’s a change in the seizure behavior, something new or different in terms of the course of the disease that's unexpected, a different type of behavior or semiology, it’s an indication to get a repeat EEG.
At the AES meeting, NeurologyLive sat down with Tatum to further explore this topic.
William Tatum, DO: My talk was on when we should obtain a routine EEG in adult patients with epilepsy. The summary of my talk was dedicated toward identifying who should have an EEG as routine test and what it will bring to the table relative to the clinician taking care of people.
The initial portion of the talk was some historical background including the discovery of the EEG by Hans Berger back in 1929, he actually recorded the EEG from his own son Klaus who was 16. Why I thought that was important was because it was fought with intense controversy and with intense skepticism compared to today where the American Academy of Neurology has shown that 55% to 60% of all practicing neurologist read EEG and have consistently done so for the last decade, so the impact is high in terms of those who actually participate in doing EEG.
When we should eat do an EEG as a routine, is really broken down into 3 different levels: diagnosis, classifying, and managing. For diagnosis an EEG is very helpful after a first seizure with Level A evidence, suggesting that if it shows epileptiform abnormalities it really is a diagnosis of epilepsy and those people need treatment. It's helpful in terms of classifying whether it's a focal or a generalized process for selecting drug treatment when somebody does present with clinical epilepsy. In managing, we can actually follow somebody with treated epilepsy by identifying routine EEGs in recurrent testing to identify whether the therapy is working. We can see whether somebody prognostically is able to be tapered off of medicine by demonstrating epileptiform abnormalities or not, to suggest that they have a higher versus a lower risk of whether they'll have a seizure recurrence and subsequently be at risk for morbidity and injury or mortality from things like a car accident. We can identify photosensitivity, we can find that an individual after epilepsy surgery may have abnormal epileptiform discharges that tell us more about the residual effects of what surgery has been expected to do relative to rendering somebody's seizure free—should they be on medicine, should they be off medicine, is there a different area that's active or not.
There are some high-yield situations and some low-yield situations that may be helpful in terms of obtaining a routine EEG. The specificity is very high when somebody has an epileptic form abnormality on the EEG, so individuals with a first seizure, individuals that are being tapered off of medication, individuals that are being followed by neurologists or an epileptologist for response to drug therapy is important, following epilepsy surgery is important as well.
Those low yield situations are situations where there is a poor spatially accessible area of the brain that can't have epileptiform abnormalities identified based on the normal routine standard recording that's done using what we call the 10-20 System of Electrode Placement, so if it's deep or if it's a small area that's involved we may miss it and we may falsely miss it. Other low-yield areas are people that complain of headache or adults that have attention deficit hyperactivity disorder or if they have an established diagnosis of epilepsy, doing another EEG is probably not going to add to that armamentarium of influencing treatment.
When we prolong or repeat the EEG we'll get increased benefit by showing that with each successive EEG there's an opportunity for greater yield, greater recovery of abnormality. Prolonging EEG the same, even recording events can be obtained with prolonged routine EEG.
We're always trying to find ways that we can influence the EEG to tell us what we want and while location is important, may or may not be near the electrodes we're recording from, age is important, the number of electrodes that we use is important, the drugs that they're on is important, the type of epilepsy they have is important, there's the number of EEGs, and the way the EEG is performed is important.
Sleep deprivation is one of the ways that we can increase our yield even beyond just obtaining sleep, hyperventilation and photic stimulation are used as a routine now and those may be helpful in people with genetic generalized epilepsy as well.
In the end relative to obtaining in a routine EEG again and again and again, if we see a change in the seizure behavior, if there's something that is new or different that we see in terms of the course that's unexpected, a different type of behavior, or semiology, that's an indication to get a repeat EEG.
The question of when has to do with high-yield situations and low-yield situations. The high-yield situations are seizures, epilepsy, and spells when they're clinically suspect; the low yield, headaches or situations where somebody already has established epilepsy.
The repeat EEGs give you an increased yield, the repeat EEGs should be done if there's a change in something along the course that is different than the stable pattern that you've otherwise seen before. Prolonging may help in terms of identifying an abnormality when you're really searching for it after a non-diagnostic EEG or trying to capture an event with/without activating procedures like hyperventilation or photic stimulation. There’re a few different scenarios to take a look at, but in the end video EEG monitoring is crucial in terms of providing a definitive diagnosis, providing a more certain classification, and characterizing an epilepsy for the purposes of epilepsy surgery.
Routine EEG is the foundation, it's not necessarily the final answer. The best advice I could give to a clinician who is going to perform an EEG is knowing what you're looking for, have a good idea anticipating whether you think somebody does or does not clinically have epilepsy because if they do have epilepsy and you believe that they do EEG should not alter that quest for ruling in or ruling out the diagnosis, and similarly, if you think they don't have epilepsy, then seeing an abnormality even though it's relatively specific for the diagnosis may not have anything to do with the diagnosis for some non-epilepsy related condition that you think they have clinically, so trust your clinical acumen.
Transcript edited for clarity