Seizures during emergence from an induced coma happen because the brain has adapted to heavy sedation and reacts with a surge of electrical excitability when that sedation is pulled back. The drugs used to maintain a medically induced coma work by dampening brain activity, and when they are reduced, the brain can overshoot into hyperexcitability. But drug withdrawal is only part of the story. The original brain injury that required the coma, shifts in body chemistry during the waking-up process, and temperature changes all contribute. The result is that seizures during the weaning phase are not a rare complication but a recognized and, in many cases, anticipated challenge that intensive care teams actively monitor for.
How Sedative Drugs Reset the Brain’s Excitability
Most drugs used to induce a medical coma work by boosting the brain’s main inhibitory signaling system. They enhance the activity of receptors that normally calm neural firing. During days or weeks of continuous sedation, the brain compensates. It dials down its own inhibitory capacity and ramps up excitatory pathways to try to restore balance. This is the same basic process that drives withdrawal symptoms from alcohol or tranquilizers, and it explains why reducing the drugs too quickly can tip the brain into seizure activity.
Research on neurosteroid withdrawal illustrates this vividly. When levels of the brain’s own calming steroid drop, the composition of inhibitory receptors physically changes. The altered receptors become less sensitive to sedative modulation, and their electrical properties shift in ways that allow more excitatory current to flow. The result is both increased anxiety and a measurably lower seizure threshold, meaning it takes less provocation to trigger a seizure.
1PubMed. Withdrawal properties of a neuroactive steroid: implications for GABA(A) receptor gene regulation in the brain and anxiety behaviorThe same principle applies on a larger scale when an ICU patient has been on continuous intravenous sedation for days. The brain has physically remodeled itself around the presence of the drug. Remove the drug, and you are left with a brain that has weakened its own braking system and strengthened the accelerator. That mismatch is the core mechanism behind withdrawal seizures.
Which Drugs Are Most Likely to Cause Problems
Not all coma-inducing drugs carry the same seizure risk during weaning, and the specifics matter because ICU teams choose their agents partly based on how manageable the withdrawal process will be.
Barbiturates, especially pentobarbital, are among the heaviest sedatives used in medically induced comas. They are often reserved for patients whose seizures have already resisted other treatments. In one cohort of patients treated with continuous pentobarbital for severe refractory seizures, the drug controlled seizure activity in the vast majority of cases, but seizures came back in roughly half of patients during the weaning process. Adding phenobarbital helped manage the rebound seizures in most of those patients.
2PubMed Central. Is pentobarbital safe and efficacious in the treatment of super-refractory status epilepticus: a cohort study A separate study examining complications of prolonged barbiturate coma found a similar pattern, with multiple patients requiring additional anticonvulsant medication because seizures returned as the barbiturate was tapered down.3PubMed Central. Observed medical and surgical complications of prolonged barbiturate coma for refractory status epilepticus
Propofol, another common choice for induced coma, has its own quirks. It has been linked to abnormal motor events that look like seizures, and the exact nature of these episodes has been debated. In at least one documented case, seizure-like activity after propofol sedation was accompanied by a biological marker (a rise in prolactin levels) that strongly suggested a true seizure rather than just muscle twitching.4PubMed. Propofol withdrawal seizures (or not) This matters because if clinicians dismiss post-propofol jerking as harmless movement, they could miss a real seizure that needs treatment.
Benzodiazepines round out the major drug classes used for induced coma. Withdrawal seizures from benzodiazepines have been documented since the 1960s and can occur with short-acting, medium-acting, and long-acting versions of the drug if they are stopped abruptly. The risk increases with higher doses and longer durations of use, both of which are typical in ICU coma protocols.5PubMed. Benzodiazepine withdrawal seizures and management
The Original Brain Injury Has Not Gone Away
It is easy to focus on the drugs and forget why the patient was put into a coma in the first place. Induced comas are not elective. They are typically used for patients with severe brain injuries, uncontrollable seizures, or brain swelling that threatens survival. The underlying damage does not heal during the coma; the coma simply buys time and reduces metabolic demand on injured tissue.
When sedation is lifted, the injured brain is exposed again. Damaged neurons in and around the injury zone may have become prone to abnormal electrical firing. Scar tissue forming after a traumatic injury or stroke can act as a seizure focus, a spot where disorganized electrical activity tends to start. So the brain waking up from an induced coma faces a double hit: drug withdrawal making it more excitable than normal, and pre-existing damage that was already generating a seizure risk before the coma began.
In patients who suffered cardiac arrest and lost oxygen to the brain (a common reason for induced coma), post-anoxic myoclonus adds another layer of complexity. Involuntary jerking movements are common after oxygen deprivation, and distinguishing between myoclonus that originates from the cortex (and may be associated with seizures) and myoclonus that originates from deeper brain structures (and usually is not) is clinically important. Research has found that cortical myoclonus occurs about twice as often as subcortical myoclonus after anoxic brain injury, and patients with cortical myoclonus have substantially higher rates of electrographic seizures.6PubMed Central. Early myoclonus following anoxic brain injury In other words, when the jerking is cortical in origin, it is often a sign of genuine seizure activity rather than a benign movement artifact.
Temperature Shifts and Electrolyte Swings
Patients in induced comas often undergo therapeutic hypothermia, where body temperature is deliberately lowered to protect the brain. The cooling phase itself tends to suppress seizure activity, but the rewarming phase is a well-known danger zone. In a study of children who received hypothermia after cardiac arrest, seizures began during the late hypothermic or rewarming periods in nearly all cases where seizures occurred.7PubMed Central. Electroencephalographic monitoring during hypothermia after pediatric cardiac arrest The rising temperature appears to unmask seizure activity that was kept in check by the cold, adding yet another trigger on top of the drug withdrawal and the underlying injury.
Electrolyte imbalances are another common culprit, and ICU patients are particularly vulnerable to them. Sodium, calcium, and magnesium levels can all shift during prolonged sedation due to intravenous fluid management, kidney function changes, and the metabolic effects of the sedative drugs themselves. Acute electrolyte disturbances, especially low sodium, low calcium, and low magnesium, can provoke seizures on their own. These are classified as acute symptomatic seizures, meaning they are caused directly by the metabolic disturbance rather than by epilepsy. The critical point is that correcting the electrolyte problem is more important than adding antiseizure medication, because the seizures will not stop until the underlying chemistry is fixed.8PubMed Central. Acute Symptomatic Seizures Caused by Electrolyte Disturbances
The Blood-Brain Barrier Problem
The blood-brain barrier is a tightly controlled gateway that normally prevents most substances in the bloodstream from reaching brain tissue. In patients with brain injuries, this barrier is often compromised. When it leaks, substances that would not normally contact neurons, including inflammatory molecules and proteins from the blood, can flood into brain tissue and trigger abnormal electrical activity. Research has pointed to blood-brain barrier dysfunction as a significant factor in seizure disorders, and some investigators have suggested that combining standard antiseizure drugs with treatments that stabilize the barrier could be more effective than antiseizure drugs alone.9PubMed Central. The etiological role of blood-brain barrier dysfunction in seizure disorders
For patients emerging from induced coma, the barrier issue is compounded. The original injury (trauma, stroke, infection, or oxygen deprivation) may have damaged the barrier. The inflammation that follows any serious brain insult keeps it leaky. And the process of rewarming, if hypothermia was used, can itself increase barrier permeability. So the brain is not just dealing with drug withdrawal and metabolic shifts; it is doing so while its own protective filter is compromised, letting in substances that further lower the seizure threshold.
Seizures You Cannot See
One of the more unsettling realities of the weaning process is that many seizures during emergence from induced coma are invisible to the naked eye. Non-convulsive status epilepticus is a condition where the brain is in continuous or near-continuous seizure activity, but the patient shows no obvious convulsions. They may simply appear unconscious or confused, which in a patient emerging from a coma can easily be mistaken for normal slow awakening. In ICU patients with impaired consciousness, non-convulsive status epilepticus has been reported in a wide range of cases, with estimates running from about 8% to 37% of unconscious ICU patients.10PubMed Central. Non-convulsive status epilepticus as a cause of delayed emergence after a thoracic surgery: a case report
This is why continuous brain-wave monitoring is standard practice during the weaning of an induced coma in many intensive care units. Without it, a patient could be seizing for hours without anyone realizing. The consequence of missed non-convulsive seizures is not just delayed waking; prolonged seizure activity can cause additional brain damage on top of whatever injury led to the coma in the first place.
How ICU Teams Try to Predict and Prevent Rebound Seizures
Given that roughly half of patients on pentobarbital coma experience seizure recurrence during weaning, predicting who will seize and who will wake up smoothly is a major focus of neurocritical care research. The standard approach is to wean sedation gradually, not abruptly, while maintaining continuous brain-wave monitoring and adjusting antiseizure medications as needed.
Recent research has identified a promising predictive tool based on the electrical patterns visible during the deepest phase of sedation. When a patient is in barbiturate or anesthetic coma, the brain-wave tracing shows a distinctive pattern called burst suppression, where brief bursts of electrical activity alternate with periods of near-silence. It turns out that not all bursts are equal. In patients who went on to have seizure recurrence during weaning, the bursts contained more fast-frequency electrical activity. In patients who weaned successfully, the bursts were dominated by slower activity. The difference was striking and statistically robust, suggesting that analyzing the content of these bursts could give clinicians an early warning about who is at high risk for rebound seizures before they even begin reducing the drug.11PubMed Central. Spectral properties of bursts in therapeutic burst suppression predict successful treatment of refractory status epilepticus
Another question that arises during weaning is how aggressively to respond when worrisome electrical patterns appear on the monitor but the patient is not overtly seizing. Some patterns sit in a gray zone between normal brain activity and clear seizure activity. Case series have explored whether clinicians can tolerate these ambiguous patterns and continue weaning rather than reflexively increasing sedation every time the brain-wave tracing looks concerning. In a small group of patients managed with a more aggressive weaning approach despite the emergence of these intermediate patterns, favorable outcomes were reported in each case, suggesting that automatically treating every borderline brain-wave finding with more sedation may unnecessarily prolong coma and its associated complications.12PubMed. Successful Wean Despite Emergence of Ictal-Interictal EEG Patterns During the Weaning of Prolonged Burst-Suppression Therapy for Super-Refractory Status Epilepticus
Why the Weaning Process Is So Slow
If you have a family member in an induced coma, the pace of weaning can feel agonizingly slow. It might take days or even weeks to fully reduce the sedation. This is deliberate. Rapid withdrawal of sedative drugs virtually guarantees a rebound in brain excitability, and in a brain that is already injured, that rebound can be catastrophic. Gradual tapering gives the brain’s inhibitory system time to partially recover from the adaptive changes it made during sedation.
During the taper, the medical team is juggling multiple variables at once. They are reducing the coma-inducing drug while simultaneously adjusting maintenance antiseizure medications, monitoring electrolytes and correcting imbalances, controlling temperature, and watching the continuous brain-wave monitor for early signs of trouble. If seizures break through, the team may need to pause the wean, increase sedation temporarily, add a new antiseizure drug, or some combination of all three. Each restart delays the emergence process further.
The difficulty is compounded by the fact that prolonged sedation itself carries serious risks, including infections, blood clots, muscle wasting, and organ dysfunction. So clinicians face a genuine tension: weaning too quickly risks seizures and additional brain damage, but weaning too slowly exposes the patient to the cumulative harms of staying in a coma. The burst-frequency research described earlier may eventually help resolve this tension by identifying patients who can safely be weaned faster, but for now, the process remains more art than algorithm.
When Jerking Movements Are Not Actually Seizures
Not everything that looks like a seizure during coma emergence is one. Patients waking from deep sedation frequently exhibit involuntary movements, including twitching, jerking, and abnormal posturing. Some of these are genuine seizures. Others are myoclonus (sudden muscle jerks) that arise from brainstem circuits rather than cortical seizure activity, and still others are simply the motor system rebooting after days of forced inactivity.
The distinction matters enormously for prognosis and treatment. As noted in research on post-cardiac-arrest patients, cortical myoclonus is associated with electrographic seizures in over 40% of cases, while subcortical myoclonus carries a much lower risk of about 8%.6PubMed Central. Early myoclonus following anoxic brain injury But the two types can look identical at the bedside. This is another reason continuous brain-wave monitoring is so valuable: it can confirm whether the movements are accompanied by seizure activity in the brain or are just involuntary motor events without an electrical seizure correlate.
For families watching at the bedside, the practical takeaway is that jerking movements during emergence are common and do not automatically mean the patient is having a seizure or getting worse. The ICU team uses the brain-wave monitor to distinguish between concerning and benign movements, and their assessment of what is happening electrically inside the brain is far more reliable than what the movements look like from the outside.
Long-Term Seizure Risk After Induced Coma
A question that often comes up once the immediate crisis has passed is whether the patient will continue to have seizures after they are fully awake and off all sedation. The answer depends heavily on the underlying cause. Patients whose induced coma was triggered by a structural brain injury, like a severe traumatic brain injury or a large stroke, carry a meaningful long-term risk of developing epilepsy. The injury itself creates lasting changes in brain tissue that can serve as a seizure focus for months or years afterward. In these cases, the seizures during coma emergence may have been the first sign of a chronic seizure disorder that will require ongoing medication.
By contrast, patients whose seizures during weaning were driven primarily by drug withdrawal or electrolyte imbalances often have a much better long-term outlook. Once the withdrawal period is over and body chemistry stabilizes, the provocation that was triggering the seizures is gone. These patients may not need long-term antiseizure medication at all, though the decision about when and whether to stop medication is highly individual and depends on factors like the severity of the underlying brain injury and whether the brain-wave monitoring remains abnormal after the patient wakes up.
The uncertainty around long-term risk is one reason that follow-up neurology care after discharge from the ICU is so important. Brain-wave studies done weeks or months later can help clarify whether a patient still has abnormal electrical activity that warrants continued treatment, or whether the brain has stabilized enough to consider tapering medications. The transition from ICU-level seizure management to outpatient epilepsy care is its own complex process, and patients and families should expect that the medication plan established during the acute hospitalization will likely be revisited and adjusted over time.