Approach to the discontinuation of antiseizure medications

INTRODUCTION — In most patients, a long durable remission from seizures can be achieved with the use of antiseizure medications (ASMs) [1]. When seizures are controlled, patients often want to consider stopping ASMs. However, this is a complex issue with a risk-benefit analysis that must be offered to each patient so that they fully understand the risks and burdens of continuing ASM therapy versus the risk of seizure recurrence with stopping ASMs. Numerous variables need to be considered so that an informed decision can be made. While more studies are needed to better address this issue, a general approach to the elimination of ASMs can be constructed for an individual patient based on their unique epilepsy and seizure characteristics.

This topic review will discuss the approach to discontinuing ASMs for patients with epilepsy who are seizure-free for several years.

IMPORTANT CONSIDERATIONS

When to consider ASM withdrawal — In most cases, it is reasonable to begin a discussion about a trial of antiseizure medication (ASM) discontinuation versus ASM continuation for adults and children who are seizure-free for at least two years. (See 'Duration of seizure freedom' below.)

The author of this topic does not consider withdrawal of ASM therapy unless at least two years of seizure freedom have elapsed, and the author favors five years of seizure absence before considering elimination of an ASM. Specific situations where consideration of ASM withdrawal is appropriate include ASM used as a prophylaxis for an intracranial neurosurgical procedure, and benign childhood-onset epilepsy syndromes; examples include childhood absence epilepsy and self-limited epilepsy with centrotemporal spikes (SeLECTS), previously termed "benign rolandic epilepsy." (See 'Epilepsy syndrome and pathology' below.)

A 2021 position paper from the American Academy of Neurology (AAN) stated that withdrawal of ASM therapy can be considered for most children who are seizure-free for at least 18 to 24 months and who do not have a high-risk epilepsy syndrome [2]. (See 'Estimating seizure recurrence risk' below.)

Individualized decision — The decision to withdraw ASM therapy for children and adults with epilepsy in remission must be individualized and weighs the risks of seizure recurrence against the possible benefits of ASM withdrawal, all of which may vary significantly across patients. The role of shared decision-making and patient counseling is paramount in this process.

Benefits of ASM withdrawal — There are several reasons to consider discontinuing ASMs in appropriate patients. Most of the potential benefits of stopping ASMs relate to an improved quality of life derived from eliminating the risks and burdens of ASM therapy [3-7].

●No drug is entirely benign. Common and rare but serious adverse effects of ASMs are listed in the tables (table 1A-B). Adverse effects of some ASMs may include bone disease, teratogenic impact on unborn children of females who take these medications, and/or complex drug interactions with nonseizure medications. Cognitive and behavioral side effects of ASMs may be subtle and not fully recognized until the medications are discontinued [3].

●Financial issues related to the cost and purchase of chronic ASMs may pose a significant burden for many patients.

●There is stigma and inconvenience attached to taking a chronic medication for epilepsy.

●There may be special circumstances, such as pregnancy or serious coexisting medical conditions, in which outcomes may be improved and management simplified in the absence of unnecessary ASM therapy.

Many patients want to eliminate ASMs and often ask when that can occur [8].

Risks of ASM withdrawal — The main risk of withdrawing and discontinuing ASM therapy is seizure recurrence. This is a particularly important consideration for individuals who are employed, who drive, and/or whose lifestyle would be adversely affected by recurrent seizures. There is also a risk of seizure-related injuries and even the remote possibility of fatal injury. Therefore, it is essential for both the patient and clinician to understand the risks surrounding safety and driving associated with seizure recurrence.

There is also a small risk that seizure control will not be regained when ASM therapy is resumed in the event of recurrent seizures. (See 'Poorly-controlled epilepsy with seizure recurrence' below.)

Seizure recurrence after stopping ASM — The seizure recurrence rate following ASM discontinuation in various studies has ranged from 12 to 67 percent [9-13]. The risk of recurrence is highest within the first year after ASM discontinuation, particularly the first six months, and decreases thereafter [9,11,14]. Thus, the prognosis tends to be more favorable if a patient does not relapse within the first year after stopping ASM. However, even patients who are seizure-free for several years and have none of the risk factors listed below (see 'Independent predictors' below) still have approximately a 20 to 25 percent risk of seizure recurrence after ASM withdrawal, a higher risk of seizures than the general population.

In a 2015 meta-analysis that included 7082 patients who attempted ASM withdrawal, the cumulative risk of seizure recurrence following ASM discontinuation at one, two, and three or four years was 22, 28, and 34 percent, respectively [11]. Other systematic reviews have suggested lower seizure recurrence rates among children (12 to 52 percent) compared with adults (46 to 66 percent) [12]. In the 1991 multicenter Medical Research Council (MRC) trial, which enrolled over 1000 patients free of seizures for at least two years, seizure recurrence at two years after randomization was 41 percent for patients assigned to ASM withdrawal compared with 22 percent for patients assigned to continued ASM treatment [13].

Poorly-controlled epilepsy with seizure recurrence — Most patients with seizure recurrence after ASM withdrawal will regain seizure control with resumption of ASM therapy. There is a small risk that the same degree of seizure control achieved before ASM discontinuation will not be regained when ASM therapy is resumed after seizure recurrence; poorly-controlled epilepsy despite resumption of ASM therapy affects ≥9 percent of patients with seizure recurrence [10,15].

However, the decision to taper is probably not responsible for the development of refractory epilepsy. One study followed 409 patients with epilepsy who, having been seizure-free for at least two years, developed seizure recurrence after being randomly assigned to ASM withdrawal or continuation in the 1991 multicenter MRC trial [13]; there was no difference between the two groups in the risk of refractory epilepsy [16].

The AAN practice advisory noted that ASM withdrawal may not increase the risk of status epilepticus in adults [2]. This conclusion was based upon the results of a single trial that randomly assigned adult patients who were seizure-free for two years on ASM monotherapy to ASM withdrawal (n = 79) or continuation (n = 81); at one year of follow-up, no patients in the withdrawal arm had status epilepticus. [3]. The AAN practice advisory did not identify any studies that addressed this question in children [2].

ESTIMATING SEIZURE RECURRENCE RISK

Variables that impact the risk — To best understand the individual's risk of seizure recurrence following antiseizure medication (ASM) withdrawal, it is important to understand how various demographic, electro-clinical, and individual factors impact the risk of recurrence. More than 20 different variables for seizure recurrence have been identified [11]. However, these risk factors have not been studied consistently across the different reports that are the basis for understanding this issue. Therefore, discussing risk factors for seizure recurrence may lead to confusion or misunderstanding when counseling patients.

A description of various risk factors and how they impact seizure recurrence following ASM withdrawal are listed in the table (table 2).

Independent predictors — In a meta-analysis of individual patient data from 10 studies and 1769 patients examining risk of seizure recurrence after ASM discontinuation, the following factors were independent predictors of seizure recurrence at two and five years after initiation of ASM withdrawal and/or seizures in the last year of follow-up [15]:

●Seizure-free interval before ASM withdrawal, with shorter interval (eg, less than two years of seizure freedom) associated with higher risk

●Epilepsy duration before remission, with longer duration associated with higher risk

●Age at onset of epilepsy, with onset in adulthood associated with higher risk (see 'Age of epilepsy onset' below)

●Number of seizures before remission, with ≥10 seizures associated with higher risk

●History of febrile seizures

●Absence of a self-limiting epilepsy syndrome (see 'Epilepsy syndrome and pathology' below)

●Developmental delay, defined by an intelligence quotient (IQ) less than 70 or by clinical judgment if IQ not available

●Epileptiform abnormality on electroencephalogram (EEG) before ASM withdrawal (see 'Role of EEG' below)

●Focal seizures (see 'Epilepsy syndrome and pathology' below)

●Number of ASMs before withdrawal (see 'Number of ASMs before withdrawal' below)

●Female sex

●Family history of epilepsy

The magnitude of risk for any one of these variables ranged from 1.3 to 1.5 times the risk of recurrence compared with the absence of the variable. A history of seizure recurrence with past attempts at drug withdrawal was not an independent predictor of future failure.

Duration of seizure freedom — ASM withdrawal in patients with less than two years of seizure freedom is associated with an increased risk of seizure recurrence [2,11,15]. Every additional seizure-free year is associated with a reduced risk [15].

Age of epilepsy onset — The age of onset has been identified as predictive in several studies and a meta-analysis with individual patient data [11,15,17]. In childhood, there is a U-shaped relationship with an elevated risk in those with epilepsy onset at birth, the lowest risk at ages three to four years, and then an increasing risk with epilepsy onset up to the age of 10 years. The risk of recurrence plateaus with epilepsy onset from 10 to 25 years [15]. With onset after 25 years, the risk of recurrence is high and continues to increase linearly.

However, it is unclear if genetic epilepsy syndromes, as well as other causes of epilepsy that are seen across different ages, have been accounted for in these analyses. As such, it is unclear whether age is a true independent risk factor for recurrence, despite the findings of the meta-analyses, in which most included studies were considered low to moderate quality [11,15].

Epilepsy syndrome and pathology — The type of epilepsy syndrome is important in helping to frame the question of whether successful remission from epilepsy off ASMs can occur.

●Favorable – There are several childhood-onset epilepsy syndromes that have favorable prognoses for seizure remission and successful ASM elimination. These include:

•Self-limited epilepsy with centrotemporal spikes (SeLECTS), previously termed "benign epilepsy with centrotemporal spikes" or "rolandic epilepsy" (see "Self-limited focal epilepsies of childhood", section on 'Self-limited epilepsy with centrotemporal spikes (SeLECTS)')

•Self-limited epilepsy with autonomic seizures (SeLEAS), previously termed "early-onset benign childhood occipital epilepsy" or "Panayiotopoulos syndrome" (see "Self-limited focal epilepsies of childhood", section on 'Self-limited epilepsy with autonomic seizures (SeLEAS)')

•Childhood absence epilepsy (see "Childhood absence epilepsy")

●Unfavorable – Epilepsy syndromes with an unfavorable prognosis for remission include [12]:

•Juvenile myoclonic epilepsy (see "Juvenile myoclonic epilepsy")

•Focal epilepsies such as temporal lobe epilepsy and frontal lobe epilepsy (see "Focal epilepsy: Causes and clinical features")

•Diffuse epilepsy syndromes such as the Lennox-Gastaut syndrome (see "Lennox-Gastaut syndrome")

In these syndromes, it is best to counsel the patient that although elimination of drug is theoretically possible, the risks are still overwhelmingly against successful withdrawal of ASMs.

Lesional epilepsies, such as those related to stroke, tumors, heterotopias, or trauma, are often associated with persistent drug-resistant epilepsy [15]. Although certain individuals may successfully find remission, especially if they undergo curative resective surgery, many will likely require lifelong ASM. Other factors that impact the recurrence risk after ASM withdrawal are the characteristics of the seizures themselves. Focal seizures are associated with an increased risk of recurrence. The presence of more than one seizure type as seen in Lennox-Gastaut syndrome is also associated with higher recurrent risk [12].

Number of ASMs before withdrawal — The number of ASMs needed to produce remission and in use at the time of ASM withdrawal is another factor associated with the risk of seizure recurrence; taking more than one ASM is associated with an increased risk [11,15].

Prior epilepsy surgery — One of the main goals of patients who undergo resective surgery is eliminating ASM therapy. However, the question of eliminating ASM therapy after epilepsy surgery is not well studied. This issue is reviewed separately. (See "Surgical treatment of epilepsy in adults", section on 'Antiseizure medication management after surgery' and "Seizures and epilepsy in children: Refractory seizures", section on 'Antiseizure medication management after surgery'.)

Others — Other factors that may increase risk but for which the data are less consistent include [3,18-23]:

●Abnormal neurologic examination

●Multiple seizure types

●Poor initial response to treatment

●Hippocampal atrophy or abnormal hippocampal signal on magnetic resonance imaging (MRI)

Neuroimaging is not well studied as a predictive factor, and there is no clear evidence that abnormal imaging impacts the risk of seizure recurrence [2]. However, MRI is an essential aspect of epilepsy management, and MRI findings may have a future role in prognosticating seizure recurrence after ASM withdrawal, especially with the higher levels of imaging strength that are now utilized in epilepsy clinical care.

Nomogram for predicting individualized risk — A nomogram for individualized prediction of seizure recurrence risk derived from the patient-level data in the 2017 individual participant meta-analysis of ASM withdrawal is available at Epilepsy prediction tools [15]. The accuracy of this tool has been suboptimal in some independent validation studies [24-26]. The nomogram should not be used as a substitute for an individualized discussion of a full range of potential risks and benefits. It does not apply to patients with acute symptomatic seizures or neonatal seizures as these populations were not included in the study, or to patients with juvenile myoclonic epilepsy, who were underrepresented and who are known to have a high rate of seizure relapse (>80 percent) with attempted ASM withdrawal. Patients who have had epilepsy surgery were also underrepresented in the study. (See "Juvenile myoclonic epilepsy", section on 'Prognosis'.)

Role of EEG — There are differing acceptable practices among experienced epileptologists concerning the role of EEG in the decision-making process for ASM withdrawal. The author obtains an ambulatory 24-hour EEG to capture wakefulness and sleep prior to stopping ASMs in most children and adults to help with risk stratification, if there is agreement to pursue consideration of ASM withdrawal. EEG provides modest prognostic information in children and may provide some prognostic information in adults. If the EEG shows any epileptiform discharges, the likelihood of successful ASM withdrawal is probably reduced. However, epileptiform EEG abnormalities should not be used as the sole reason to decide against ASM withdrawal in the absence of other factors associated with an increased risk of seizure recurrence [15]. (See 'Independent predictors' above.)

While EEG abnormalities are generally associated with higher seizure recurrence risk, the available evidence supports the association of an epileptiform EEG with an increased risk mainly in children; the evidence for adults is limited [2]. This is likely due to the technologic and methodologic variability often seen in these studies. In addition, interictal EEG activity can be suppressed by ASM therapy.

A 2021 American Academy of Neurology (AAN) practice advisory concluded that interictal epileptiform activity on EEG possibly increases the risk of seizure recurrence in children [2] and cited one small study in children that found a higher risk of seizure recurrence with interictal epileptiform activity on EEG compared with no interictal epileptiform activity (24 of 51 [47 percent] versus 31 of 94 [33 percent], odds ratio 2.87, 95% CI 1.35-6.11) [27]. However, the practice advisory found insufficient evidence in adults due to a paucity of EEG studies that address this question [2]. The practice advisory concluded that it is unknown if EEG informs the decision to withdraw ASMs.

By contrast, a 2017 systematic review and meta-analysis with individual patient data from children (n = 1474) and adults (n = 295) found that an epileptiform abnormality on EEG before ASM withdrawal was associated with an increased risk of seizure recurrence compared with a normal EEG (hazard ratio 1.50, 95% CI 1.25-1.79), but an abnormal EEG only slightly increased the risk in the absence of other clinical risk factors [15]. Another 2017 meta-analysis of trials with 2349 patients with epilepsy also found that an abnormal EEG was associated with an increased risk of seizure recurrence (pooled risk ratio 1.37, 95% CI 1.20-1.55) [28].

SHARED DECISION-MAKING AND COUNSELING — Patients generally want to account for their individual goals and values when making decisions about their care. Shared decision-making incorporates the patient's values and preferences into medical decisions and puts the patient at the center of care. When technical evidence about the relative merits of different treatment decisions is equivocal, the patient's perceptions about the nontechnical aspects of care are particularly relevant to the medical decision.

Thus, clinicians should engage patients and their families and caregivers throughout the decision-making process of antiseizure medication (ASM) discontinuation by exploring the patient’s values and preferences, discussing overall benefits and risks of ASM withdrawal, and incorporating individual variables to estimate a patient's specific risk of seizure recurrence. (See 'Benefits of ASM withdrawal' above and 'Risks of ASM withdrawal' above and 'Estimating seizure recurrence risk' above.)

●Patients should be counseled that the highest risk of recurrence is during the first year after discontinuation. They should also be counseled about the potential (low) risk of status epilepticus or death [2] and instructed what to do in the event of a seizure. The patient should be comfortable accepting these risks prior to proceeding with ASM discontinuation.

●Patients' motivations and expectations should be explored and discussed to make sure that they are realistic with respect to potential outcomes. It is worth discussing which specific problems patients hope to improve (eg, avoiding ASM side effects, expense, burden, or other issues) to get a better sense of whether ASM withdrawal is right for them.

●For adults, it is essential to discuss what the impact would be if they were unable to drive due to a seizure recurrence. In the event of a seizure, patients could lose driving privileges according to laws in their jurisdiction. Moreover, since the risk of recurrence is higher during drug tapering, patients should be advised not to drive during this period. (See "Driving restrictions for patients with seizures and epilepsy", section on 'Discontinuing medication'.)

Like driving, there may be some careers or jobs (eg, operating heavy machinery, working at heights, piloting an aircraft) in which seizure recurrence may raise concerns with public safety limitations. Similarly, seizure recurrence may impact personal safety or ability to function with attending school, activities of daily living (eg, bathing), and recreational activities (eg, swimming). These issues are just as important as the other clinical factors (eg, type of epilepsy, number of ASMs used, EEG findings) and must be considered alongside the individual patient’s values and preferences.

RATE OF ASM WITHDRAWAL — For patients who wish to purse antiseizure medication (ASM) discontinuation, the ASM should be tapered rather than halted abruptly. For patients who take more than one ASM, only one ASM should be tapered and discontinued at a time.

It is probably best to taper slowly to hedge against the risk of seizure recurrence. The taper schedule varies by type of ASM. Benzodiazepines and barbiturates are associated with withdrawal seizures and should be discontinued very gradually. The author tapers benzodiazepine and phenobarbital slowly over a period of six months or more, while the author tapers other ASMs over a period of three to four months.

Rapid tapering (over days to a few weeks) may increase the risk of seizures, but there is debate about the optimal ASM taper rate, as there are only limited data to guide an ASM tapering schedule [29-31]. One study in children found no clear difference in overall seizure recurrence risk at three years between rapid (six-week) or slow (nine-month) taper of ASM, but quicker tapers relapsed sooner [29].

MONITORING AFTER ASM WITHDRAWAL — Several studies have examined the role of EEG monitoring during antiseizure medication (ASM) discontinuation [19,32-37]. A 2019 meta-analysis identified seven studies with 703 patients with at least two years of seizure freedom and reported that abnormal EEG findings (defined as epileptiform abnormalities in five studies, and epileptiform and nonepileptiform abnormalities in two studies) during taper were associated with higher seizure recurrence risk (57 versus 32 percent, odds ratio 3.84, 95% CI 2.03-7.27) [33]. An earlier study performed serial EEGs during ASM withdrawal to guide resumption of ASM therapy in a subset of patients if increasing epileptiform activity was seen; this strategy was associated with a seizure relapse rate of 23 percent, compared with 43 percent in patients who continued ASM withdrawal in the presence of increasing EEG epileptiform activity [38].

Although there is no consensus regarding monitoring, the author often obtains a 24-hour ambulatory EEG within one month of ASM discontinuation and suggests resuming ASM therapy if epileptiform discharges are detected. Otherwise, in the absence of epileptiform activity on EEG or recurrence of seizures, the patient can continue off ASM without need for ongoing follow-up or monitoring.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Seizures and epilepsy in adults" and "Society guideline links: Seizures and epilepsy in children".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Seizures (The Basics)" and "Patient education: Epilepsy in adults (The Basics)" and "Patient education: Epilepsy and pregnancy (The Basics)" and "Patient education: Epilepsy in children (The Basics)")

●Beyond the Basics topics (see "Patient education: Seizures in adults (Beyond the Basics)" and "Patient education: Seizures in children (Beyond the Basics)" and "Patient education: Treatment of seizures in children (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●When to consider ASM withdrawal – For adults and children who are seizure-free for at least two years, it is reasonable to begin a discussion about antiseizure medication (ASM) continuation versus a trial of discontinuation. The decision to withdraw ASM therapy must be individualized and weighs the risks of seizure recurrence against the possible benefits of ASM withdrawal. Shared decision-making and patient counseling is paramount in this process. (See 'When to consider ASM withdrawal' above and 'Individualized decision' above.)

●Benefits and risks of stopping ASM therapy – Most of the potential benefits of stopping ASMs relate to an improved quality of life derived from eliminating the risks and burdens of ASM therapy. The main risk of withdrawing and discontinuing ASM therapy is seizure recurrence. There is also a small risk that seizure control will not be regained when ASM therapy is resumed for recurrent seizures. (See 'Benefits of ASM withdrawal' above and 'Risks of ASM withdrawal' above.)

●Predictors of seizure recurrence – Independent predictors of seizure recurrence after ASM discontinuation include the following (see 'Independent predictors' above):

•Seizure-free interval before ASM withdrawal, with shorter interval (eg, ≤2 years seizure freedom) associated with higher risk

•Epilepsy duration before remission, with longer duration associated with higher risk

•Age at onset of epilepsy, with onset in adulthood associated with higher risk (see 'Age of epilepsy onset' above)

•Number of seizures before remission, with ≥10 seizures associated with higher risk

•History of febrile seizures

•Absence of a self-limiting epilepsy syndrome (see 'Epilepsy syndrome and pathology' above)

•Developmental delay, defined by an intelligence quotient (IQ) less than 70 or by clinical judgment if IQ not available

•Epileptiform abnormality on EEG before ASM withdrawal (see 'Role of EEG' above)

•Focal seizures (see 'Epilepsy syndrome and pathology' above)

•Number of ASMs before withdrawal (see 'Number of ASMs before withdrawal' above)

•Female sex

•Family history of epilepsy

●EEG – We obtain an EEG prior to stopping ASMs in most children and adults to help with risk stratification. However, epileptiform EEG abnormalities should not be used as the sole reason to decide against ASM withdrawal in the absence of other factors associated with an increased risk of seizure recurrence. (See 'Role of EEG' above.)

●Shared decision-making – Clinicians should engage patients, families, and caregivers throughout the decision-making process of ASM discontinuation by exploring the patient’s values and preferences, discussing overall benefits and risks of ASM withdrawal, and incorporating individual variables to estimate a patient's specific risk of seizure recurrence.

It is essential to discuss the impact that seizure recurrence might have on the patient's lifestyle related to driving, work, school, activities of daily living, and recreational activities. (See 'Shared decision-making and counseling' above.)

●Tapering ASM – For patients who wish to pursue ASM discontinuation, the ASM should be tapered (slowly) rather than halted abruptly. For patients who take more than one ASM, only one ASM should be tapered and discontinued at a time. (See 'Rate of ASM withdrawal' above.)