false
Catalog
SCCM Resource Library
Decoding the EEG: All That Spikes Does Not Seize
Decoding the EEG: All That Spikes Does Not Seize
Back to course
[Please upgrade your browser to play this video content]
Video Transcription
Thank you for having me and thank you for staying for the last talk of the session. So coming back to our case, during setup for optic nerve sheet diameter assessment, the patient was noted to have a forced gaze deviation along with eyelid and facial twitching. And so my talk is going to focus on when to consider EEG monitoring in our patients and how to approach treatment based on EEG findings. These are my disclosures, none directly related to this talk. And so just to start with, there's been an increasing trend in the use of continuous EEG monitoring in the critical care setting, not just for brain injury and not just in the neuro-ICU setting but also outside of the neuro-ICU for patients with altered mental status in sepsis, hepatic encephalopathy, with increasing recognition that many of these patients are at high risk for having subclinical seizures. And so when should we consider EEG monitoring? The American Clinical Neurophysiology Society has published guidelines and recommendations on when to consider EEG monitoring in critically ill or ICU patients. The most common indication and the most frequent indication is for persistently altered mental status after generalized convulsive seizures. Typically after generalized convulsive seizures, a patient should have returned in their mental status over 30 minutes to an hour. Persistent altered mental status beyond that point, so if you've had a patient who's had generalized seizures, is not waking up at 30 minutes to an hour, you should consider EEG monitoring. Up to 90% of non-convulsive seizures are actually detected within the first 24 hours for patients who have non-convulsive seizures after generalized status. In patients not presenting with seizures but have altered mental status, we should consider EEG. If they have a structural brain injury plus altered mental status, especially if the mental status is out of proportion to what their imaging shows. So if they have a right MCA stroke and their exam shows left-sided weakness and aphasia, I wouldn't jump to EEG monitoring, but if they have a small frontal contusion and their exam shows their mental status is way out of proportion to what the frontal contusion would explain or deficits that don't match up with the frontal contusion and no other reason to explain it, I would consider EEG monitoring. Unexplained altered mental status or fluctuating mental status without CNS injury, particularly in our hepatic encephalopathy or sepsis patients that we know are at a high risk for seizures. Other indications for EEG monitoring if there is a need for paralysis and a high risk for seizures, so acute brain injury patients that also have ARDS, for instance, and need to be paralyzed for that or have high ICP and need sedation for ICP management. Patients that require sedation for monitoring depth of sedation, so this is refractory status patients or again, refractory intracranial pressure patients where we're using sedation to control seizures or ICP. EEG is used as a guide. Paroxysmal events, this could be recurrent motor symptoms, sensory phenomenon, autonomic phenomenon that cannot be explained by any other etiology, we can consider EEG monitoring for these patients. Prognostication, we can use EEG for prognostication in TBI, hypoxic ischemic injury after cardiac arrest and subarachnoid hemorrhage. Out of all three of these, most commonly EEG is used in prognostication after cardiac arrest patients who have anoxic brain injury. Ischemia detection, there's an increasing role of EEG in DCI or delayed cerebral ischemia detection in subarachnoid hemorrhage patients and we often use it in our high-grade subarachnoid hemorrhage patients at our center. So what are the other aspects? So once we identify a patient that we think needs EEG, when should we start monitoring and how long should we monitor these patients? And majority, as I said, majority of patients that have non-convulsive seizures after statuses are recognized very early during the recording and almost all are recognized within the first 24 hours. Of course, not everyone has access to continuous EEG monitoring or may only have access to brief EEG monitoring. And so even the initial findings on the EEG, if there's seizures or epileptiform activity early in the EEG, these are the patients we'd want to consider monitoring for longer. If the initial EEG is reassuring, there's normal background or normal architecture or no concerning findings, these are the patients that we can monitor for briefer or shorter periods of time or even consider intermittent EEG monitoring. How often should the EEG be reviewed? Ideally, continuously by a provider that's continuously monitoring the EEG, which of course is not always possible. And so the recommendation is that there should be at least twice a day review and report with more frequent review depending on how active the EEG is. And simultaneous video monitoring is essential in these patients. So what happens? What do we do with our EEG findings? The common finding is electrographic seizures, which is what we were expecting or suspecting in our patient. And how do we treat electrographic seizures or status? There's no large trials that have looked at this comprehensively, but we typically draw from the recommendations that have been published for convulsive or generalized convulsive status epilepticus. The first line approach is typically benzodiazepines. Many of our patients are already in the ICU, either mechanically ventilated or we... Benzodiazepines are typically first response either in the field or as soon as they hit the emergency. And so often this stage is bypassed and we go straight to the first line anti-seizure medications. And here we pick medications based on the patient's profile and risk profile and contraindications. Typically levotiracetam is the agent of choice because it tends to interact less with all the other medications our patients are on and has less and is not impacted by hepatic failure. And so these are typically the first line choices and again guided by what the patient comorbidity is. A large trial that compared each one of these found no difference in seizure control across all medications and similar adverse effect profiles. So we select based on the patient profile. And if this first tier medications are not working, then we consider additional anti-seizure medications and there is a whole slew of medications that we can pick from. Typically those that can be administered IV if possible, lacosamide, topiramate, all of these medications have been looked at in smaller studies and trials. And of course if there's refractory seizures or the patient is already intubated consideration or if not and they have refractory seizures consideration for intubating mechanical ventilation and then using anesthetics. How we treat these patients with anesthetics? Again there's no large trials comparing anesthetic regimens. The consensus sort of is targeting some sort of seizure or burst suppression for at least 24 hours to 48 hours and then gradually weaning off the sedation. What we find more often and if you may have seen many of your EEG reports is more and more we're seeing that it's not seizures that are reported but there is periodic or rhythmic patterns, lateralized periodic discharges, generalized periodic discharges that are coming to be known as interictal continuum patterns. These patterns, I've given a few examples, they essentially don't meet the diagnostic criteria for seizures. Seizures on EEG is a clear start, clear end, there's evolution over space and evolution in morphology and they're very discrete events that we can identify. These periodic and rhythmic or ictal interictal continuum patterns don't have that clear cut signature. They're either sporadic or they don't evolve, they don't have a discrete onset or end. Seizures do and there's a lot of work going on on what is the diagnostic, what is the prognostic significance and there are many questions that are still unanswered as to are these ictal, are they interictal, what is their relation to seizures, do they result in secondary brain injury, do they impact outcomes and therefore that sort of leads challenges on how to approach and treat them. I'll do a quick review of some of the small studies that have looked at these patterns. Essentially what we found is that these patterns are associated with a higher risk for seizures, especially if they're lateralized periodic discharges or generalized periodic discharges that are high frequency and high burden. We know seizures result in secondary brain injury and a sympathetic overdrive systemically and so many small cohorts have tried to see if they can reproduce these findings in association with these patterns. Pet studies have shown periodic and rhythmic patterns are associated with pet hypermetabolism similar to seizures suggesting potential secondary brain injury. This was a study in subarachnoid hemorrhage patients that showed periodic patterns were associated with significant reductions in brain tissue oxygenation, again suggesting potential secondary brain injury and impact on outcomes. Traumatic brain injury, these periodic discharges have been shown to be associated with drops in brain glucose, rise in brain lactate pyruvate ratios, again suggesting secondary brain injury. There are studies that show these patterns are associated with a higher risk for long-term epilepsy. And we've looked at burden of these patterns and impact on outcomes and found that as the quantitative burden, so the more time spent in these periodic and rhythmic patterns, the higher the likelihood of poor discharge as well as long-term functional outcomes. And this holds true across disease categories, brain injury, hypoxic ischemic encephalopathy, status epilepticus, and systemic illnesses such as sepsis or hepatic encephalopathy, respiratory disorders in the absence of any acute structural brain injury. So how do we, you know, they're associated with seizures, they're associated with secondary brain injury, they're associated with higher risk for outcomes. How do we treat them? Should we be treating them? And unfortunately, that's, this is ongoing work. There is no clear answer on to what, you know, or large randomized trials to tell us whether treatment improves outcomes. And in the absence of clear-cut data, seizure medications are frequently started in response to these patterns with widespread practice variation, some providers being more aggressive, others being less aggressive. And seizure medications are not benign, they have their own adverse effects and can adversely impact outcomes in this patient population. So in the absence of large trials and, you know, and as studies are ongoing, how do we approach these patterns? And this tends to be, you know, a general approach is first and foremost, is there a clinical correlate? If there is a clear clinical correlate, consider treating or consider an anti-seizure medication. Determine if there is a significant metabolic derangement, if a patient has generalized or diffuse periodic discharges and their BUN is 500 or, not 500, but they have uremia or they have a clear hepatic encephalopathy, then, you know, you want to focus on also correcting that metabolic derangement. Determine if they had seizures prior to EEG, if they had seizures, then these patterns are more likely to be associated with further risk for seizure. And then finally, what is the pattern type and frequency, is it high burden, is it high frequency, are there lateralized periodic discharges or lateralized rhythmic delta, which actually increase, which are more likely to be associated with seizures. So we sort of keep all of this in mind in deciding our treatment approach. There is some data that ancillary testing, such as MRI or PET scans, can also help in management. These patterns can be associated with a diffusion restriction in MRI, suggesting there's secondary brain injury and you may want to consider treatment. PET scans, if these patterns are, this is a patient who has periodic discharges, they had a PET scan that showed hypermetabolism in that same region, and so that guided treatment in this patient using invasive monitoring. So this is a traumatic brain injury patient where on the scalp recording they had rhythmic delta or irregular, at some points irregular delta slowing, but this patient also had a depth, a deep EEG intracranial electrode and we saw clear periodic discharges and seizures in relation to these patterns and therefore treated based on these findings. And so this is sort of, again, no large trials that have done this yet, but a lot, consensus approach tends to be if you have these periodic and rhythmic patterns, if there's a clear clinical correlate, consider a treatment trial. If there's no clear clinical correlate but they are high frequency, high burden, and in the clinical context you can consider treatment trial. If there's no clinical correlate, they're low frequency, low burden, they're not occupying the entire EEG, then you can continue to monitor if there's metabolic derangements, aggressively treat metabolic derangements, and then if you see continued worsening of the clinical status or EEG, you can initiate treatment. And the treatment, again, is whether you use anti-seizure medications or you go to anesthetics or use benzos, it sort of depends on the patient's clinical status and comorbidities. A patient that is already intubated, mechanically ventilated, you may consider a more aggressive treatment approach. Patients where you're trying to avoid intubation or trying to avoid over sedation, you may want to use less sedating medications. And once we start treatment, you monitor clinically, see if there's clinical improvement, monitor the EEG, know that there could be a lag between EEG improvement and clinical improvement. And of course, if we're not seeing any changes or improvement before throwing on five other anti-seizure medications, consider other approaches, consider ancillary testing if that is available. So to summarize, unexplained mental status that can't be explained by imaging or any other clinical factors, consider EEG monitoring. And the treatment approach should really take into consideration what is happening clinically and treat in the clinical context. Thank you for your time.
Video Summary
The speaker discussed the use of EEG monitoring in patients with altered mental status in the critical care setting. They outlined various indications for EEG monitoring, including persistently altered mental status after seizures, unexplained altered mental status, and fluctuating mental status without CNS injury. They also discussed the use of EEG for prognostication, detection of ischemia, and the management of periodic and rhythmic patterns. The speaker highlighted the need for clinical correlation and consideration of metabolic derangements when deciding on treatment approaches for these patterns. Overall, EEG monitoring plays a crucial role in the assessment and management of patients with altered mental status.
Asset Subtitle
Neurosceince, Procedures, 2023
Asset Caption
Type: one-hour concurrent | Windows to the Brain: Neuromonitoring for the General Intensivist (SessionID 1202529)
Meta Tag
Content Type
Presentation
Knowledge Area
Procedures
Knowledge Area
Neuroscience
Membership Level
Professional
Membership Level
Select
Tag
electroencephalogram EEG
Tag
Seizure
Year
2023
Keywords
EEG monitoring
altered mental status
critical care
indications
prognostication
Society of Critical Care Medicine
500 Midway Drive
Mount Prospect,
IL 60056 USA
Phone: +1 847 827-6888
Fax: +1 847 439-7226
Email:
support@sccm.org
Contact Us
About SCCM
Newsroom
Advertising & Sponsorship
DONATE
MySCCM
LearnICU
Patients & Families
Surviving Sepsis Campaign
Critical Care Societies Collaborative
GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. |
Privacy Statement
|
Terms & Conditions
The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.
×
Please select your language
1
English