ASM Withdrawal & Discontinuation
The question of when and whether to discontinue antiseizure medications (ASMs) is one of the most consequential decisions in epilepsy management. After achieving seizure freedom, patients understandably wish to stop taking medication, and clinicians must weigh the desire for a medication-free life against the risk of seizure recurrence and its potentially devastating consequences—including injury, loss of driving privileges, psychosocial harm, and rarely sudden unexpected death in epilepsy (SUDEP). The decision requires careful consideration of the epilepsy syndrome, EEG findings, duration of seizure freedom, etiology, and the specific ASM being withdrawn. Certain epilepsy syndromes have predictably high remission rates (childhood absence epilepsy, benign epilepsy with centrotemporal spikes), while others carry a high relapse risk when ASMs are discontinued (juvenile myoclonic epilepsy). This topic reviews the evidence and guidelines for ASM withdrawal, practical taper protocols, and the implications of withdrawal for driving and daily life.
Bottom Line
- When to consider withdrawal: Generally after ≥2 years of seizure freedom in adults (≥1–2 years in children); individual risk-benefit assessment is essential
- Recurrence risk: Overall 30–40% risk of seizure recurrence after ASM withdrawal in adults; highest risk in the first 12 months (especially the first 6 months); most recurrences happen within 2 years
- Risk factors for recurrence: Epileptiform EEG at time of withdrawal, structural brain lesion, adolescent/adult onset, longer duration of active epilepsy before control, multiple seizure types, history of drug-resistant epilepsy
- Safe to withdraw: Childhood absence epilepsy (remission >70%), benign epilepsy with centrotemporal spikes (BECTS, nearly 100% remission), and childhood epilepsy syndromes with known age-limited course
- Risky to withdraw: Juvenile myoclonic epilepsy (relapse rate >80–90% after withdrawal), epilepsy with structural lesion, epilepsy with persistent EEG abnormalities
- Taper protocol: Gradual reduction over at least 2–3 months per ASM; withdraw one drug at a time in polytherapy; some ASMs (carbamazepine, oxcarbazepine, benzodiazepines) require particularly slow tapers
When to Consider ASM Withdrawal
General Criteria
The question of ASM withdrawal arises most commonly in two settings: (1) children who have outgrown an age-limited epilepsy syndrome, and (2) adults who have been seizure-free for an extended period. The AAN practice parameter (2004) and subsequent guidance suggest considering withdrawal under the following circumstances:
- Seizure-free duration: At least 2 years in adults; at least 1–2 years in children (shorter intervals may be appropriate for age-limited syndromes)
- Normal neurologic examination: No focal neurologic deficits suggesting an ongoing structural process
- EEG at time of withdrawal: Normal or improved (absence of new epileptiform discharges); abnormal EEG at withdrawal is the strongest predictor of recurrence
- Patient preference: After thorough discussion of recurrence risk and consequences (driving, employment, safety)
The Risk-Benefit Conversation
Before initiating ASM withdrawal, clinicians must discuss:
- Benefits of withdrawal: Elimination of medication side effects (cognitive, metabolic, teratogenic), reduced cost, improved quality of life, psychological freedom from the "epilepsy patient" identity, avoidance of long-term ASM toxicities (bone density, hepatotoxicity)
- Risks of recurrence: Seizure-related injury (falls, burns, drowning), loss of driving privileges (in most US states, 3–12 months seizure-free required to drive), psychosocial consequences (employment, independence), risk of status epilepticus (rare), SUDEP (very rare but may occur with GTC seizures)
- Timing considerations: Avoid withdrawal during periods of high seizure risk (pregnancy, sleep deprivation, major life stress); plan withdrawal when consequences of a breakthrough seizure are minimized
Factors Favoring vs. Opposing ASM Withdrawal
- Favoring withdrawal:
- Age-limited epilepsy syndrome (BECTS, childhood absence epilepsy)
- Seizure-free ≥2–5 years
- Normal EEG at time of withdrawal decision
- No structural brain lesion
- Single seizure type
- Idiopathic/genetic etiology
- Seizures controlled on first ASM (monotherapy)
- Significant ASM side effects affecting quality of life
- Teratogenic ASM in woman planning pregnancy
- Opposing withdrawal:
- Juvenile myoclonic epilepsy (JME)
- Epileptiform EEG at time of withdrawal consideration
- Structural brain lesion (cortical dysplasia, tumor, hippocampal sclerosis)
- Adolescent or adult onset
- History of drug-resistant epilepsy (multiple ASM failures before control)
- Multiple seizure types
- Occupation requiring alertness (commercial driver, heavy machinery, pilot)
- History of severe consequences from prior seizures
Risk Factors for Seizure Recurrence After Withdrawal
| Risk Factor | Effect on Recurrence Risk | Evidence Level |
|---|---|---|
| Epileptiform EEG at withdrawal | Strongest predictor; recurrence risk approximately doubled (up to 60–70%) | Multiple prospective studies; meta-analyses |
| Structural brain lesion | Significantly increased risk; lesional epilepsy often requires lifelong therapy | Observational studies |
| Adolescent/adult onset (vs. childhood) | Higher recurrence risk in adult-onset epilepsy | Prospective cohorts; meta-analysis |
| Duration of active epilepsy before control | Longer duration of uncontrolled epilepsy before seizure freedom predicts higher recurrence | Observational studies |
| Multiple seizure types | Higher recurrence than single seizure type | Meta-analysis |
| Polytherapy at time of withdrawal | Higher recurrence than monotherapy (implies more refractory epilepsy) | Observational studies |
| JME diagnosis | Relapse rate >80–90% after withdrawal—generally considered lifelong treatment | Multiple prospective studies |
| Shorter seizure-free interval (<2 years) | Higher recurrence than longer seizure-free intervals | Randomized trials; meta-analysis |
| Family history of epilepsy | Modest increase in recurrence risk | Inconsistent evidence |
Epilepsy Syndromes and Withdrawal Prognosis
| Syndrome | Withdrawal Feasibility | Expected Relapse Rate | Recommended Approach |
|---|---|---|---|
| BECTS (benign epilepsy with centrotemporal spikes) | Excellent | <5% (nearly all remit by age 14–16) | Withdraw after 1–2 years seizure-free or by mid-adolescence; may not even require treatment initially |
| Childhood absence epilepsy (CAE) | Good | 20–30% (higher if GTC seizures develop or if there is incomplete response to initial therapy) | Attempt withdrawal after 2 years seizure-free; EEG should be normal; higher relapse if evolves to JME |
| Juvenile absence epilepsy (JAE) | Moderate | 40–50% | Higher relapse than CAE; consider long-term low-dose therapy; individualize based on EEG |
| Juvenile myoclonic epilepsy (JME) | Poor | >80–90% | Generally considered lifelong treatment; withdrawal almost invariably leads to relapse; if attempted, do so very gradually with close monitoring |
| Focal epilepsy (unknown cause, normal MRI) | Moderate | 30–40% | Consider after ≥2–5 years seizure-free; normal EEG at withdrawal; discuss risk-benefit |
| Focal epilepsy (structural lesion) | Poor | 50–70% | Withdrawal generally not recommended unless the lesion has been surgically resected and the patient has been seizure-free for ≥2 years postoperatively with normal EEG |
| Post-surgical epilepsy (temporal lobectomy) | Moderate | 20–40% (varies by study and surgical outcome) | Consider after ≥2 years seizure-free postoperatively; multidisciplinary decision; normal postoperative EEG favors success |
Juvenile Myoclonic Epilepsy: The Lifelong Treatment Paradigm
- JME has the highest relapse rate of any common epilepsy syndrome after ASM withdrawal (>80–90%)
- Relapses often occur within weeks to months of drug withdrawal, frequently with generalized tonic-clonic seizures
- Even after decades of seizure freedom on medication, withdrawal typically leads to recurrence
- The concept of "growing out of" JME is largely a myth—the electrographic and clinical tendency persists
- Lifelong low-dose monotherapy (typically valproate in males, lamotrigine or levetiracetam in females) is the standard recommendation
- Rare patients may successfully withdraw, but this should be the exception and requires careful individualized assessment
Taper Protocols
General Principles
- One drug at a time: In patients on polytherapy, withdraw only one ASM at a time; complete the taper and observe for ≥3–6 months before attempting to withdraw the next
- Gradual reduction: A general rule is to reduce the dose by 25% every 2–4 weeks, completing the taper over at least 2–3 months; faster tapers increase the risk of withdrawal seizures
- Monitor clinically: Patient should report any auras, myoclonic jerks, or changes in seizure pattern; obtain EEG if recurrence is suspected
- Timing: Begin withdrawal when consequences of a breakthrough seizure are minimized (e.g., not immediately before a driving test, job interview, or pregnancy)
ASM-Specific Taper Considerations
| ASM | Taper Considerations | Recommended Taper Duration |
|---|---|---|
| Carbamazepine / Oxcarbazepine | High risk of rebound seizures with abrupt withdrawal; generalized tonic-clonic seizures may occur even in patients with focal epilepsy; must taper slowly | ≥3 months; reduce by ~200 mg (CBZ) or ~300 mg (OXC) every 2–4 weeks |
| Benzodiazepines (clobazam, clonazepam) | Significant withdrawal risk: withdrawal seizures, anxiety, insomnia, and rarely status epilepticus; physical dependence develops with chronic use | ≥3–6 months; very slow taper with small decrements (e.g., 5 mg clobazam or 0.25 mg clonazepam per 2–4 weeks) |
| Phenobarbital | Long half-life (80–100 hours) provides some self-tapering effect; still requires gradual withdrawal due to dependence and withdrawal seizure risk | ≥3 months; reduce by 15–30 mg every 2–4 weeks |
| Valproate | Moderate withdrawal risk; taper over 2–3 months; watch for emergence of myoclonic or absence seizures in IGE | 2–3 months; reduce by 250 mg every 2 weeks |
| Lamotrigine | Generally well tolerated during taper; if co-administered with valproate, remember valproate doubles lamotrigine levels—reducing valproate will lower lamotrigine levels | 2–3 months; reduce by 50 mg every 2 weeks |
| Levetiracetam | Relatively low withdrawal seizure risk; post hoc data suggest efficacy is maximal at initial dose, so taper may be straightforward | 2–4 weeks; reduce by 500 mg every 1–2 weeks |
| Perampanel | Very long half-life (~105 hours); can be discontinued abruptly without need for taper | No taper necessary |
| Lacosamide | Standard taper; moderate risk | 2–3 months; reduce by 50–100 mg every 2 weeks |
| Topiramate / Zonisamide | Standard taper; moderate risk; watch for rebound headache in migraine patients | 2–3 months; reduce by 25–50 mg (TPM) or 100 mg (ZNS) every 2 weeks |
Withdrawal Seizures vs. Epilepsy Recurrence
It is clinically important to distinguish between withdrawal seizures and true epilepsy recurrence, as the implications for subsequent management are different:
| Feature | Withdrawal Seizure | Epilepsy Recurrence |
|---|---|---|
| Timing | During active taper or within days to weeks of dose reduction/discontinuation | Weeks to months after complete withdrawal; may occur even after extended seizure-free interval off medication |
| Seizure type | Often generalized tonic-clonic regardless of the patient's usual seizure type; particularly with carbamazepine, oxcarbazepine, and benzodiazepine withdrawal | Typically the patient's habitual seizure type |
| Mechanism | Loss of pharmacologic suppression; rebound neuronal hyperexcitability; physical dependence (benzodiazepines, barbiturates) | Return of underlying epileptic tendency; unmasking of ongoing epileptogenesis |
| Management | Slow the taper or temporarily increase the dose; resume previous effective dose and reattempt slower taper later (if appropriate) | Resume ASM therapy; may need to return to previous regimen or consider alternative if the patient had been seizure-free for a prolonged period |
| Prognosis | Does not necessarily indicate ongoing need for lifelong ASM therapy; a slower taper may succeed | Indicates persistence of the epileptic tendency; most patients will require long-term ASM therapy |
Driving Implications
Driving During and After ASM Withdrawal
- Most US states require 3–12 months of seizure freedom before allowing driving (varies by jurisdiction)
- ASM withdrawal creates a period of increased seizure risk, even before a seizure has occurred
- AAN recommendation: Patients should be advised not to drive during the withdrawal period and for at least 3 months after the last dose reduction (some clinicians recommend 6 months)
- If a seizure occurs during withdrawal, the clock restarts for the mandatory seizure-free driving interval
- Commercial driving licenses (CDL) have more stringent requirements; most states require extended seizure-free intervals and documentation of ASM stability
- Patients should be counseled about driving restrictions before initiating withdrawal—many patients reconsider withdrawal when they understand the driving implications
- Document the driving discussion in the medical record
Post-Surgical ASM Withdrawal
After successful epilepsy surgery (typically temporal lobectomy or lesionectomy), many patients and clinicians consider ASM withdrawal. The evidence is summarized below:
- Engel Class I outcomes (seizure-free): ASM withdrawal can be considered after ≥2 years of postoperative seizure freedom
- Risk factors for postoperative recurrence: Incomplete resection, persistent postoperative epileptiform EEG, dual pathology, extratemporal surgery, preoperative generalized seizures, older age at surgery
- Success rates: Approximately 60–80% of patients who attempt postoperative withdrawal remain seizure-free; the majority who relapse regain seizure freedom after ASM reinstitution
- Practical approach: Obtain postoperative EEG; if normal and seizure-free ≥2 years, discuss withdrawal with the patient; taper gradually (one drug at a time); if on multiple ASMs, withdraw the adjunctive agent first while maintaining the primary ASM
When Recurrence Occurs
Managing Seizure Recurrence After Withdrawal
- Reinstate the previously effective ASM at the dose that maintained seizure freedom; most patients (80–90%) regain seizure control after ASM reinstitution
- If the recurrence occurred during active taper: resume the previous dose and reassess whether withdrawal is appropriate
- If the patient had been off ASMs for months before recurrence: this represents true epilepsy recurrence and long-term therapy is likely needed
- Obtain an EEG to assess for epileptiform activity and guide treatment decisions
- Reassess driving eligibility based on local regulations; restart the mandatory seizure-free interval
- Discuss whether another withdrawal attempt will be considered in the future; if the first attempt failed, the risk of recurrence with a second attempt is higher
AAN/ILAE Guideline Summary
Key recommendations from the AAN practice parameter on discontinuing ASMs (2004) and subsequent ILAE guidance:
- Children who have been seizure-free for 2 years may be offered ASM withdrawal, with the understanding that ~30% will have a seizure recurrence
- Adults who have been seizure-free for ≥2 years can consider ASM withdrawal after individualized risk-benefit assessment
- An abnormal EEG at the time of withdrawal is the strongest predictor of recurrence and should weigh heavily against withdrawal
- Taper should be gradual (minimum 2–3 months; longer for benzodiazepines and barbiturates)
- The decision should be shared between clinician and patient, with clear documentation of the discussion, including driving implications
- Most recurrences happen within the first 12 months, especially the first 6 months; if seizure-free at 2 years post-withdrawal, the long-term outlook is favorable
References
- Abou-Khalil B. Update on antiseizure medications 2025. Continuum (Minneap Minn) 2025;31(1):123–165.
- Specchio LM, Beghi E. Should antiepileptic drugs be withdrawn in seizure-free patients? CNS Drugs 2004;18(4):201–212.
- Beghi E, Giussani G, Grosso S, et al. Withdrawal of antiepileptic drugs: guidelines of the Italian League Against Epilepsy. Epilepsia 2013;54(Suppl 7):2–12.
- Braun KPJ, Schmidt D. Stopping antiepileptic drugs in seizure-free patients. Curr Opin Neurol 2014;27(2):219–226.
- Practice parameter: a guideline for discontinuing antiepileptic drugs in seizure-free patients. Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 1996;47(2):600–602.
- Lossius MI, Hessen E, Mowinckel P, et al. Consequences of antiepileptic drug withdrawal: a randomized, double-blind study (Akershus Study). Epilepsia 2008;49(3):455–463.
- Schmidt D, Löscher W. Uncontrolled epilepsy following discontinuation of antiepileptic drugs in seizure-free patients: a review of current clinical experience. Acta Neurol Scand 2005;111(5):291–300.
- Malow BA, Blaxton TA, Stertz B, Theodore WH. Carbamazepine withdrawal: effects of taper rate on seizure frequency. Neurology 1993;43(11):2280–2284.
- Azar NJ, Wright AT, Wang L, Song Y, Abou-Khalil BW. Generalized tonic-clonic seizures after acute oxcarbazepine withdrawal. Neurology 2008;70(22 Pt 2):2187–2188.
- Camfield P, Camfield C. When is it safe to discontinue AED treatment? Epilepsia 2008;49(Suppl 9):25–30.