So we're going to do this, and we're going to do brain death, and then it'll be lunchtime. Most brain wave activity should not be easily recognizable as coordinated. It should be generally uncoordinated. A seizure is coordinated brain activity that leads to rhythmic brain discharges. That's typically not normal. Epilepsy is recurrent, unprovoked seizures. EEGs for general seizures typically have beginning, middle, and end, where you can see that EEG is not epileptogenic. You see rhythmicity starting, and then it generalizes, spreads, and then eventually the seizure terminates, and they stop seizing. You may have heard these terms at some point during neurology rotation. Many of you have repressed this. That's probably OK. Typical findings for seizures, they're rhythmic. They often have a peak. They're often asymmetric. Generalized seizures, like thalamic seizures and absence seizures, that's only in kids. Grown-ups don't have absence seizures. There are a variety of ratios to help you distinguish when patients are having seizures. If you go down to your epilepsy monitoring unit, where they're often also monitoring ICU patients, it looks like mission control. And they're all staring at all these screens all the time, looking at the squiggly lines, and often responding to these sorts of alerts to try and help them detect patients who are having seizures, more likely to have seizures, and if someone is known to have seizures, if they are still, in fact, seizing. There's a variety of seizures that are clearly seizures. And there's a variety of EEG descriptions that are possibly seizures, which is on what the epileptology buddies call the ictal-interictal continuum, which means you have to keep asking what they mean. And I think that they're Star Trek fans. About 10% of patients in the MICU, if you go around and randomly monitor comatose patients, will be in non-convulsive status. So the more you screen for subclinical seizures, the more you find. When patients are seizing, their outcome is much, much worse. In neuro-ICUs, if you have blood in your head and altered consciousness, you should be monitored. In general ICU setting, you should probably talk to your intensives about that. And they will help you triage who ought to be monitored or not. The usual answer is, if someone is not waking up and they really ought to be, we'll probably recommend you monitor that person for a day or two, because there's a reasonable chance they'll be seizing. Status epilepticus means the seizures don't stop. This is a medical emergency. How bad the status is and how long it went depends upon the etiology. If you don't stop the seizures, the patient never wakes up. If you stop the seizures, they sometimes wake up depending on the etiology. Meaning, if you have malignant status after cardiac arrest, somebody had a schemey time of 35 minutes and the entire cortex is infarcted, the status is a surrogate marker of the severity of the cardiac arrest. That patient's highly unlikely to regain consciousness. If the patient had a brain tumor section, went in status epilepticus, but the rest of the brain is normal, if you stop the status, most of those patients will wake up a day or two or three later. They've changed the definition of status a couple of times for the person of this group. If the patient doesn't stop seizing in five minutes, you're going to call it status. And prolonged status will lead to Dane bramage, especially in the mesial temporal lobe, where seizures often terminate. And they will often be hyper intense. The mortality of status depends upon the underlying etiology, as we discussed. You want to stop status as soon as you can. There are a variety of trials of best seizure medications to do this. Giveaway the answer is lorazepam. I'm sorry for those that hate benzos. This is one time where they actually work. And then after that, it's all downhill in terms of evidence base. In general, you want to stop status with lorazepam. If that doesn't work, then you move on to phenytoin or phosphenytoin. And if that doesn't work, then this is going to be a tough day. And it may or may not go well. And you may end up using high dose barbiturates or high dose anesthetic. You know it's bad status when you've moved to so much IV midazolam. The pharmacist says, you're using up all the midazolam in the hospital. Should I tell the operating rooms to be ready to use something else tomorrow because you're going to run out? And the answer is, yes, tell them to use something else. I'm going to use all the midazolam in the hospital. I think our current record is like 135 milligrams an hour to terminate someone's status. Do that for a day, you'll use up all of the big hospitals midazolam for that week until they get another shipment in. This happens a couple of times a year. Why would you do that? Well, some patients with status wake up. Young women with NMDA receptor encephalitis will seize for months until you essentially carpet bomb the immune system's suppression or you find a teratoma. You stop the immunologic process that's leading to seizures. And young women that have literally been in status for months wake up and go back to work. In terms of prospective randomized clinical trials, these have typically been very highly selected patients with a very clear endpoint. In general, lorazepam is better than everything else. Once you get past the lorazepam, it doesn't really matter. There is a, for patients that have been resistant to benzodiazepines, there's a trial of phosphonatone versus high-dose levotiracetam. There is really no difference in successful stopping of seizures after that. But this protocol of 60 milligrams per kilogram levotiracetam load caught on. And so for most patients with status, we're using this high-dose levotiracetam load. It doesn't seem to lead to hypotension. And it seems to be as effective as anything else. If you don't have this, then valproate or phosphonatone is fine. You don't have to put the patient in burst suppression to control refractory status. It doesn't mean that the patient won't have more seizures. But it's easy to train people to see this overnight when they're on call in the ICU. And people have all seen this, or you may have seen this. There's a group of epileptologists somewhere. I'm convinced they're watching Star Trek when we all leave. But they say they're looking at all these monitors all the time, monitoring what's in the ICU, and we go back and forth. Any questions about seizures and status before I talk about brain death? Short version, don't give prophylactic seizure meds to patients with blood in their head, except for TBI. It's part of the guidelines for TBI, but for spontaneous brain damage, don't give prophylactic seizure meds. Guidelines say don't do it. It makes the patients do worse. Well-intentioned, reasonable, doesn't work.

seizures

epilepsy

status epilepticus

EEG

treatment