Initial treatment of epilepsy in adults

Initial treatment of epilepsy in adults

Literature review current through: Jan 2022. | This topic last updated: Jul 19, 2021.


epilepsy is the syndrome of two or more unprovoked seizures that occur more than 24 hours apart [1]. Seizures affect people in many different ways. Seizures are disruptive in the lives of patients and can cause injury. People with epilepsy have higher rates of psychiatric comorbidity and may experience adverse psychosocial outcomes. Most worrisome is that people with epilepsy have an approximately threefold increased mortality compared with people who do not have seizures [2]. (See “Comorbidities and complications of epilepsy in adults”.)

The management of patients with epilepsy is focused on three main goals: controlling seizures, avoiding or minimizing treatment side effects, and maintaining or restoring quality of life. The initial treatment of epilepsy is with a single antiseizure medication (ie, monotherapy). There is an ever-expanding list of available antiseizure medications. Unfortunately, studies have not identified a single antiseizure medication that is clearly superior in terms of efficacy or tolerability. Because of this, clinicians must individualize the choice of antiseizure medication for each patient.

This topic will discuss the approach to the initial treatment of epilepsy. Other topics discuss the evaluation of patients with seizures and epilepsy, other aspects of epilepsy therapy, and features of specific antiseizure medications. (See “Evaluation and management of the first seizure in adults” and “Overview of the management of epilepsy in adults” and “Evaluation and management of drug-resistant epilepsy” and “Antiseizure medications: Mechanism of action, pharmacology, and adverse effects”.)


First-time unprovoked seizure — The term unprovoked seizure refers to a seizure of unknown etiology as well as one that occurs in relation to a preexisting brain lesion or progressive nervous system disorder (often referred to as a remote symptomatic seizure). Unprovoked seizures are distinct from provoked seizures: provoked seizures are due to an acute condition such as a toxic or metabolic disturbance, head trauma, or acute stroke (ie, acute symptomatic seizures).

The decision of whether or not to start antiseizure medication therapy at the time of a first unprovoked seizure in an adult should be individualized. The main factors to consider in making the decision are:

●The risk for recurrent seizures, which varies based on clinical factors discussed below (see ‘Risk of seizure recurrence’ below)

●The approximate benefit that can be expected from immediate antiseizure medication therapy on the risk of recurrent seizure (see ‘Benefit of early versus deferred treatment’ below)

●The side effect profiles of various antiseizure medication options, which vary based on individual patient comorbidities and age (see ‘Side effect profiles’ below and ‘Comorbid medical conditions’ below)

●Patient values and preferences, particularly with regard to the social consequences of a recurrent seizure (eg, implications for driving or employment) (see ‘Benefit of early versus deferred treatment’ below)

An evidence-based guideline of the American Academy of Neurology and the American Epilepsy Society on the management of an unprovoked first seizure in adults also advocates for an individualized approach that weighs the risk of seizure recurrence against the adverse effects of antiseizure medications and considers educated patient preferences [3]. The guideline offers the following specific recommendations:

●Adults with an unprovoked first seizure should be informed that their seizure recurrence risk is greatest early within the first two years (21 to 45 percent).

●Clinical variables associated with an increased risk may include a prior brain insult, an electroencephalogram (EEG) with epileptiform abnormalities, a significant brain imaging abnormality, and a nocturnal seizure.

●Immediate antiseizure medication therapy, as compared with delay of treatment pending a second seizure, is likely to reduce recurrence risk within the first two years but may not improve quality of life. Over a longer term (>3 years), immediate antiseizure medication treatment is unlikely to improve prognosis as measured by sustained seizure remission.

●Patients should be advised that the risk of antiseizure medication adverse events may range from 7 to 31 percent and that these adverse events are likely predominantly mild and reversible.

In patients with a first unprovoked seizure who are found to have a central nervous system (CNS) abnormality on neuroimaging (such as a brain tumor or scar tissue from an old head injury or CNS infection), the risk of seizure recurrence is high. In this instance, most clinicians would start treatment after the first unprovoked seizure. In fact, such patients likely have a sufficiently high risk of seizure recurrence to meet criteria for epilepsy according to International League Against Epilepsy (ILAE) guidelines [1]. These criteria now consider patients with a single unprovoked seizure and an estimated risk of recurrence ≥60 percent over ten years to have epilepsy, similar to those with two unprovoked seizures occurring >24 hours apart. (See “Evaluation and management of the first seizure in adults” and ‘Risk of seizure recurrence’ below.)

In contrast are patients with a first unprovoked seizure who have a normal (or nonfocal) examination and normal (or nonspecific) neuroimaging. In these patients, the risk of seizure recurrence is lower, and antiseizure medication therapy may be reasonably deferred until after a second unprovoked seizure.

Patient concerns also weigh heavily in treatment decisions. If the risk of seizure recurrence is low, and the individual places a high value on avoidance of side effects, antiseizure medication therapy may be delayed. In contrast, there are some individuals who will be very concerned about seizure recurrence. In this instance, an antiseizure medication may be initiated to reduce seizure recurrence, despite what may be a low likelihood of additional seizures.

Risk of seizure recurrence — In prospective, randomized trials of individuals with a first unprovoked seizure, the estimated two-year recurrence risk in untreated patients ranges from 40 to 50 percent [4-6]. The risk of recurrence is highest in the first year after the seizure and diminishes with time; 80 to 90 percent of patients who have recurrent seizures do so within two years [7,8].

The most replicated clinical factors associated with an increased risk for seizure recurrence after a first unprovoked seizure include [4-7,9-11]:

●Epileptiform abnormalities on EEG (see “Electroencephalography (EEG) in the diagnosis of seizures and epilepsy”)

●Remote symptomatic cause, as identified by clinical history or neuroimaging (eg, brain tumor, brain malformation, head injury with loss of consciousness, prior central nervous system infection, or scarring from a prior brain injury or brain surgery)

●Abnormal neurologic examination, including focal findings and intellectual disability

●A first seizure that occurs during sleep (ie, a nocturnal seizure)

Each of these factors has been associated with an approximately 2- to 2.5-fold increased risk for seizure recurrence. The studies that identified these factors were done in mixed groups of patients; some were on medication (treatment) and some were untreated. However, there is a lack of information to guide clinicians about how these risk factors interact [3].

Other potential risk factors for seizure recurrence have been investigated and remain more uncertain. As an example, patients who have a first presentation with status epilepticus or with multiple seizures within a single day are more likely to be treated with antiseizure medications than are those with a single short-duration seizure. However, limited data suggest that presentation with status epilepticus, in the absence of other risk factors, does not increase the risk of seizure recurrence [6,7,9,12]. Similarly, whether a history of prior febrile seizures is associated with an increased risk of seizure recurrence after a first unprovoked afebrile seizure is uncertain [6,7,9,11].

Study results have conflicted as to whether a family history of epilepsy impacts recurrence risk [5-7,9,11]. This varies according to the epilepsy syndrome, since several epilepsy syndromes have been identified as monogenetic in origin.

Benefit of early versus deferred treatment — For adults presenting with an unprovoked first seizure, immediate antiseizure medication treatment reduces the risk of seizure recurrence by about 35 percent over the next one to two years [4,5,10,13-17]. This estimate is derived from a meta-analysis of five randomized trials (n = 1600 patients) comparing immediate versus delayed antiseizure medication therapy in adults with an unprovoked first seizure [3].

However, studies suggest that starting an antiseizure medication has little impact on long-term outcome. At four and five years after the first seizure, patients have similar rates of complete seizure remission whether antiseizure medication treatment was initiated immediately after the first seizure or deferred until a second seizure occurred [4,5,10,13,17]. At least one randomized trial found that 20-year mortality was not impacted by immediate versus deferred treatment [18].

In the aggregate, quality of life outcomes, as measured in one randomized study, were not different with early versus deferred treatment [19]. However, the questionnaires showed that there was a tradeoff between the adverse effects of seizures versus adverse effects of taking antiseizure medications. These findings suggest that individual patient preferences should be considered when deciding between early versus delayed treatment of a single unprovoked seizure. As an example, patients randomized to early antiseizure medication treatment were more likely to be able to drive than patients whose treatment was deferred. A need to drive or operate heavy machinery along with other occupational and psychological consequences of suffering a recurrent seizure are important considerations when deciding whether to start antiseizure medication therapy.

Second unprovoked seizure — Patients presenting with a second unprovoked seizure should be started on antiseizure medication therapy, since seizure recurrence indicates that the patient has a substantially increased risk for additional seizures (ie, epilepsy) [6,8].

In one prospective case series, the risk of another seizure after two unprovoked seizures was 73 percent at four years (most of these patients were treated with antiseizure medications) [8]. In many cases, a careful history may reveal that certain seizure types such as typical absence, myoclonic, simple or complex partial have been recurrent at the time of presentation [7].

Acute symptomatic seizure — Acute symptomatic seizures have a lower risk for subsequent epilepsy compared with remote symptomatic seizures [20]. Early management decisions, including whether or not to start an antiseizure medication, depend upon multiple factors, including the severity of the underlying illness, the cause and duration of the seizure, the expected risk of early recurrence, and the risks associated with a recurrent seizure. (See “Evaluation and management of the first seizure in adults”, section on ‘Acute symptomatic seizures’.)

Patients with seizures that occur in the setting of acute severe neurologic illness or injury (eg, stroke, traumatic brain injury, meningitis, anoxic encephalopathy) are often treated with antiseizure medications in the acute setting because of the risk of prolonged recurrent seizures or aggravation of a systemic injury. (See “Overview of the management of epilepsy in adults”, section on ‘Poststroke seizures’ and “Posttraumatic seizures and epilepsy”, section on ‘Early seizures’ and “Spontaneous intracerebral hemorrhage: Acute treatment and prognosis”, section on ‘Seizure management’ and “Spontaneous intracerebral hemorrhage: Acute treatment and prognosis”.)

A subset of acute symptomatic seizures is those that occur in the setting of an acute medical illness or metabolic disturbance (table 1). In contrast to the setting of an acute stroke or traumatic brain injury, patients with seizures provoked by metabolic derangements are generally not felt to be at risk for future epilepsy, but they are at risk for seizure recurrence in the acute setting [21]. Short-term antiseizure medication therapy may be indicated if the metabolic disturbance is expected to persist or if the initial seizure is prolonged (as in the instance of status epilepticus). (See “Evaluation and management of the first seizure in adults”, section on ‘Acute symptomatic seizures’ and “Evaluation and management of the first seizure in adults”, section on ‘When to start antiseizure medication therapy’.)

SELECTION OF AN ANTISEIZURE MEDICATIONEpilepsy is initially treated with antiseizure medication monotherapy. Almost half of patients will become seizure-free with their first antiseizure medication trial [22,23].

In choosing an initial therapy, clinicians must weigh relative efficacy and potential for adverse effects of each drug. Comparative efficacy and tolerability data are limited. Comparison trials that have been performed have not shown significant differences among various drugs in terms of efficacy. Clinicians must therefore formulate treatment plans based upon a combination of drug, seizure, and patient-specific factors.

Drug-related considerations — Aspects of antiseizure medication therapy that are relevant to drug selection include efficacy, pharmacokinetics, adverse effects, and cost.

Comparative efficacy — No single antiseizure medication is clearly the most effective or best tolerated, and there are now over 25 antiseizure medications approved for treatment of seizures in adults and/or children (table 2 and table 3). (See “Antiseizure medications: Mechanism of action, pharmacology, and adverse effects”.)

Randomized trials assessing efficacy and tolerability provide the least biased evidence of efficacy. However, these typically compare active therapy to a subtherapeutic dose of the same agent and/or to placebo rather than to an effective dose of another antiseizure medication [24]. Another limitation of most randomized trials in epilepsy is that these are usually performed testing new antiseizure medications as add-on treatment in patients with treatment-resistant illness (see ‘FDA indications’ below). Such patients may not be representative of general clinical population.

There have been a limited number of randomized trials comparing various antiseizure medications head-to-head as initial monotherapy in adults, all of which have shown similar efficacy between drugs [25,26]:

Carbamazepine versus phenytoin [27]

Phenytoin versus valproate [28]

Gabapentin versus carbamazepine [29] or pregabalin [30]

Lamotrigine versus carbamazepine [31,32] or phenytoin [33] or gabapentin [34] or pregabalin [35]

Topiramate versus valproate and carbamazepine [36] or phenytoin [37]

Oxcarbazepine versus phenytoin [38,39] or valproate [40] or carbamazepine [41]

Zonisamide versus carbamazepine [42,43]

Levetiracetam versus carbamazepine or valproate [44-46]

Lacosamide versus carbamazepine [47]

Carbamazepine extended-release versus levetiracetamzonisamidelacosamide, or eslicarbazepine [48]

Although these trials have not shown significant differences between antiseizure medications, the quality of the data remains limited by the fact that they were generally of short duration (24 or 48 weeks). Such studies can compare the incidence of short-term side effects between drugs, but they have limited power to assess relative efficacy. In general, but with some exceptions, the newer antiseizure medications are superior with respect to tolerability [26,49].

Meta-analyses of randomized trials can potentially overcome some of the limitations of individual trials, but even these studies can be problematic, since patient populations and drug doses often vary between trials; this substantively limits the ability to compare the studied treatments. In general, such studies have lacked power either to refute or substantively confirm results of individual trials [50-54].

The largest individual randomized trials examining different antiseizure medications as monotherapy for the initial treatment of epilepsy were the Standard and New Antiepileptic Drugs (SANAD) trials [55-57]. The first SANAD trials included 1721 patients with focal epilepsy and 716 patients with generalized and unclassifiable epilepsy; the SANAD II trials included 990 patients with focal epilepsy and 520 patients with generalized or unclassifiable epilepsy [58,59]. In an effort to balance methodologic rigor and practicality, the trials were not blinded [60]. The treating physician determined how quickly to titrate the medication, instead of following a standardized blinded protocol. This approach may have better approximated the “real life” use of these drugs than would a blinded trial. Outcome measures were time to treatment failure (for either inadequate seizure control or intolerable side effects) and time to achievement of a 12-month seizure remission. The main findings were:

●For patients treated for focal epilepsy, the first SANAD trail found that lamotrigine and oxcarbazepine had the longest time to treatment failure compared with carbamazepinegabapentin, and topiramate [56]. Lamotrigine and carbamazepine were associated with the shortest times to 12-month seizure remission. In the SANAD II trial, by intention-to-treat analysis, levetiracetam did not meet noninferiority criteria compared with lamotrigine for time to 12-month seizure remission and was inferior for time to treatment failure for any reason; in contrast, zonisamide did meet the criteria for noninferiority compared with lamotrigine for time to 12-month seizure remission, but was also inferior for time to treatment failure [58]. Both levetiracetam and zonisamide were more likely to fail than lamotrigine due to adverse reactions, but not because of inadequate control of seizures.  

●For patients treated for generalized and unclassifiable epilepsy, the first SANAD trial found that valproate and lamotrigine were superior to topiramate in regard to time to treatment failure [57]. For time to 12-month seizure remission, valproate and topiramate were more efficacious compared with lamotrigine. In the SANAD II trial, by intention-to-treat analysis, levetiracetam did not meet noninferiority criteria compared with valproate for time to 12-month seizure remission [59].

●In the first SANAD trials, quality of life outcomes were largely similar across treatment groups over a two-year period and did not show a clear advantage for any specific drug [61]. The strongest predictor of improved quality of life outcomes was achievement of a 12-month seizure remission.

The investigators concluded that lamotrigine should be considered the drug of first choice for focal epilepsy and valproate for generalized epilepsy [56-59]. Because the SANAD trial were unblinded, however, there was potential for bias. Also, they provided only sparse data regarding the potential of rare, often idiosyncratic, serious adverse events (eg, potential for teratogenicity with valproate). These results also do not account for other patient-specific preferences regarding the likelihood of different side effects, need for drug monitoring, potential for drug interactions, and dosing frequency [62].

Pharmacokinetics — Important pharmacologic features of individual antiseizure medications are summarized in the table and reviewed in more detail separately (table 2). (See “Antiseizure medications: Mechanism of action, pharmacology, and adverse effects”.)

Some of the more important considerations when choosing a first-line antiseizure medication include the following:

Dosing frequency – The half-lives of antiseizure medications vary considerably (table 2). For many individuals, the frequency with which a drug must be taken is an important factor in compliance (ie, adherence) and/or seizure control [63]. Optimal dose frequency for individual drugs can vary between patients.

Most antiseizure medications are prescribed in two daily doses. Antiseizure medications that often require more frequent dosing include immediate-release carbamazepinetiagabine, regular and delayed-release valproategabapentin, and pregabalin. Once daily dosing may be possible with phenobarbitalphenytoin, extended-release valproate, zonisamideeslicarbazepineperampanel, and extended-release formulations of levetiracetam and lamotrigine.

Drug interactions – The selection of an antiseizure medication should consider other prescribed medications for potential drug interactions. Clinicians should review each item on a patient’s medication list for potential drug interactions [64,65]. Specific interactions of antiseizure medications with other medications may be determined using the Lexicomp drug interactions tool.

In general, antiseizure medications with hepatic enzyme induction or inhibitory properties have the greatest potential for interactions. Enzyme induction occurs with all older antiseizure medications (phenytoinphenobarbitalcarbamazepine) except valproate and ethosuximide. Enzyme induction also occurs with a few of the more recently approved antiseizure medications such as felbamatetopiramate, and oxcarbazepine (table 2) [60].

Antiseizure medications that are hepatic-enzyme inducers increase the metabolism of other medications that are broken down by the same pathway. As an example, phenytoin induces the metabolism of warfarin, potentially leading to subtherapeutic international normalized ratio (INR) and/or an increased dose requirement of warfarin. Commonly prescribed drugs with the potential to interact with enzyme-inducing antiseizure medications include statins, calcium channel blockers, serotonin reuptake inhibitors, antipsychotics, tricyclic antidepressants, hormonal contraceptive therapy, warfarin, and many anticancer drugs [66].

In contrast, valproate is a hepatic enzyme inhibitor and may cause significant increases in serum concentrations of medications that are metabolized in the liver. Similarly, stiripentol, which is approved for Dravet syndrome, is a potent hepatic enzyme inhibitor. It is known to increase levels of the main metabolite of carbamazepine (the 10,11 epoxide) [67].

Other drug interactions relate to protein binding. Addition of a drug that is highly protein-bound will displace another protein-bound drug, increasing its free fraction. In the setting of reduced serum albumin, this effect is amplified. Hormones can also affect the levels of some antiseizure medications. For instance, lamotrigine concentrations are reduced by estrogen-containing hormonal contraceptives. (See ‘Hormonal contraception’ below.)

Aging – Antiseizure medication use in older adult patients is complicated by several factors, including age-related alterations in protein binding, reduced hepatic metabolism, and diminished renal clearance of medications. In addition, polypharmacy is more often a concern in older adults. These and other factors related to antiseizure medication selection in older adults are discussed separately. (See “Seizures and epilepsy in older adults: Treatment and prognosis”.)

Side effect profiles — The adverse effects of antiseizure medications make a significant contribution to reduced quality of life in individuals with epilepsy [68]. While many antiseizure medication side effects (eg, drowsiness, dizziness, diplopia, and imbalance) seem to be common to this entire class of medicines, others are more specific to an individual drug. These should be considered in selecting an antiseizure medication since certain side effects are either more likely or more problematic in certain patients.

Common neurotoxic and systemic side effects are summarized in the table (table 4). Less common, often idiosyncratic, but potentially serious adverse events are summarized separately (table 5).

Neurocognitive side effects – Most antiseizure medications are associated with a negative impact on cognition, but some are more problematic than others [69]. Among the older antiseizure medications, studies suggest that phenobarbital is associated with greater impairments compared with carbamazepinevalproate, and phenytoin, which have similar, but more modest negative effects [70,71]. Among the newer antiseizure medications, gabapentin and lamotrigine have been found to be less problematic than carbamazepine in their effects on cognition. Negative cognitive effects are similar with oxcarbazepine and carbamazepine [72]. Finally, a significant minority of patients taking topiramate discontinue the drug because of clinically apparent cognitive difficulties. In direct comparison studies, cognitive profiles in patients taking topiramate were worse than those taking valproate, lamotrigine, or gabapentin [71].

Hypersensitivity reactions – Stevens-Johnson syndrome (SJS), toxic epidermal necrolysis (TEN), and drug rash with eosinophilia and systemic symptoms (DRESS) are rare but severe idiosyncratic reactions, characterized by fever and mucocutaneous lesions. SJS and TEN have been most often associated with the use of carbamazepineoxcarbazepinephenytoinlamotrigine, and phenobarbital (table 5), and less commonly with valproate and topiramate; however, they have been described with almost all antiseizure medications [73,74].

The period of highest risk is within the first two months of use. For carbamazepineoxcarbazepine, and possibly phenytoin, the risk is higher in patients with the HLA-B*1502 allele, which occurs almost exclusively in patients of Asian ancestry, including South Asian Indians. The US Food and Drug Administration (FDA) recommends screening such patients for the HLA-B*1502 allele prior to starting carbamazepine and oxcarbazepine. This is discussed in more detail separately. (See “Stevens-Johnson syndrome and toxic epidermal necrolysis: Pathogenesis, clinical manifestations, and diagnosis”, section on ‘HLA types’ and “Antiseizure medications: Mechanism of action, pharmacology, and adverse effects”, section on ‘Carbamazepine’ and “Antiseizure medications: Mechanism of action, pharmacology, and adverse effects”, section on ‘Oxcarbazepine’.)

Suicidality – Antiseizure medications as a class have been associated with an approximately twofold increased relative risk of suicidal behavior or ideation based on pooled analyses of placebo-controlled trials (0.43 versus 0.22 percent) [75]. Some experts advise screening for depression at diagnosis of epilepsy and at each follow-up visit [76]. This is discussed in more detail separately. (See “Overview of the management of epilepsy in adults”, section on ‘Specific adverse reactions’ and “Comorbidities and complications of epilepsy in adults”, section on ‘Psychiatric disorders’.)

Weight gain or loss – Weight gain is associated with valproategabapentincarbamazepinevigabatrinpregabalin, and perampanel. Weight loss has been reported with felbamatetopiramate, and zonisamide.

FDA indications — The FDA indication for use of antiseizure medications may influence physician prescribing habits. For many newer antiseizure medications, the FDA indications are based on studies that showed effectiveness as add-on treatment in persons with refractory epilepsy. Monotherapy trials are performed infrequently because most physicians and patients are reluctant to be treated with placebo when effective treatments exist [24]. As a result, many antiseizure medications are not FDA-approved as initial monotherapy. Some physicians may be initially reluctant to use newer antiseizure medications as monotherapy in the initial treatment of epilepsy pending advice from colleagues and/or published expert opinion.

A historical-controlled treatment discontinuation trial design is being increasingly used in an attempt to evaluate the effectiveness of various antiseizure medications as conversion to monotherapy in patients with drug-resistant epilepsy, since randomizing patients with active seizures to a placebo conversion presents ethical concerns. The primary outcome measure is the predicted exit percentage, defined as the proportion of patients meeting a seizure-related exit criterion (eg, withdrawal due to inadequate seizure control or adverse effects) at four months, compared with a pooled historical benchmark of 65 percent (compiled from older trials that did use a placebo or subtherapeutic antiseizure medication dose) [77]. Drugs shown to be more effective than this historical benchmark in prospective trials include lamotrigine extended-release [78], levetiracetam extended-release [79], pregabalin [80], lacosamide [81], and eslicarbazepine [82]. The exit rate in these trials has ranged from 20 to 40 percent at four months.

Cost of medications — For many patients, the cost of their medication is also an issue and whether a specific antiseizure medication is on a list of preferred medications approved by a third party payer may also be influential in the choice of antiseizure medication. When cost is taken into account, for areas of the world and for individual patients with restricted resources, phenobarbital may be the treatment of choice for partial epilepsy [83]. Generic substitution can lower cost of many antiseizure medications but is occasionally associated with a change in seizure control or tolerability, although the magnitude of this risk has been debated. (See ‘Generic substitutions’ below.)

Seizure-related considerations

Focal versus generalized epilepsy — In selecting an antiseizure medication for a patient with new onset epilepsy, it is important to differentiate between a focal versus generalized epilepsy syndrome [84]. Antiseizure medications are classified as either broad or narrow spectrum agents (table 3). While broad spectrum agents treat both focal and generalized epilepsy syndromes, narrow spectrum agents treat one or the other [85].

Most of the narrow spectrum agents are effective for localization-related or focal epilepsies. As an example, gabapentin (a narrow spectrum agent) may work well for a patient with temporal lobe epilepsy (a focal epilepsy), but is unlikely to be effective in juvenile myoclonic epilepsy (a generalized epilepsy). Ethosuximide is another narrow spectrum agent used for absence seizures (a generalized epilepsy), which is generally ineffective for focal seizures. Broad spectrum agents are effective for both types of epilepsies [86]. If the clinician is unsure whether the epilepsy syndrome is focal or generalized, a broad spectrum agent is usually chosen (table 3).

Identifying the correct epilepsy syndrome is critical to selecting an optimal treatment. For instance, a few of the narrow spectrum agents have been reported to worsen certain seizures that occur in the primary generalized epilepsy syndromes. Oxcarbazepine [87], carbamazepinephenytoinvigabatrin, and gabapentin [86] have all been reported to worsen certain seizures types in generalized epilepsy syndromes.

Specific etiologies — In addition to the distinction between generalized and focal epilepsy discussed above, specific etiologies of epilepsy may impact the treatment choice.

●Post-stroke epilepsy is generally easily controlled with antiseizure medication monotherapy. The choice of antiseizure medication may be influenced by specific concerns, such as potential impact of the antiseizure medication on post-stroke functional recovery and the potential for drug interactions with warfarin and salicylates (see ‘Pharmacokinetics’ above) [88]. The treatment of post-stroke epilepsy is discussed in detail separately. (See “Overview of the management of epilepsy in adults”, section on ‘Poststroke seizures’.)

●Brain tumors are associated with epilepsy in 30 to 70 percent of patients, depending on the tumor type. The choice of antiseizure medication in this setting is influenced by potential drug interactions with chemotherapeutic agents leading to decreased efficacy of both treatments, as well as the increased potential for allergic cutaneous reactions when antiseizure medications are used during radiotherapy [89,90]. Both of these factors contribute to a strong preference for non-enzyme inducing antiseizure medications in brain tumor patients when possible. (See “Seizures in patients with primary and metastatic brain tumors”.)

Comorbid medical conditions — Medical comorbidities are important to consider when selecting an antiseizure medication. Many antiseizure medications are either metabolized by the liver, excreted by the kidneys, or both (table 2). When a person has hepatic or renal disease, it may be necessary to avoid certain antiseizure medications or to adjust the dose. Other comorbidities can be problematic because of potential drug side effects or drug interactions, while others may represent an opportunity to choose an antiseizure medication that has efficacy in both conditions. Specific interactions of antiseizure medications with other medications may be determined using the Lexicomp drug interactions tool.

Renal disease — Renally excreted drugs include gabapentintopiramatezonisamidelacosamidelevetiracetamoxcarbazepine, and pregabalin (table 2) [85,91,92]. The dose of these drugs should be adjusted based on the severity of renal impairment (table 6).

In patients on hemodialysis, antiseizure medication regimens should be individualized based on drug levels and clinical response. The renally excreted drugs and some others (eg, phenobarbitallamotrigine) are removed by hemodialysis, and a low dose should be supplemented after dialysis to maintain therapeutic levels. The effects of peritoneal dialysis on antiseizure medication metabolism are not well studied, and antiseizure medication treatment in such patients may require additional monitoring. (See “Seizures in patients undergoing hemodialysis”, section on ‘Dosing’.)

Highly protein-bound antiseizure medications exhibit altered pharmacokinetics, including greater therapeutic and toxic effects and drug interactions, when given in usual doses to patients with low serum albumin or protein-binding affinity (eg, due to nephrotic syndrome or acidosis). Albuminuria (causing low serum albumin) and acidosis reduce protein binding fractions and binding affinity, leading to increased fractions of free drug [91]. For highly protein-bound antiseizure medications (table 2), subtherapeutic total drug levels may be both sufficient for efficacy and required to avoid toxicity in this setting. Free drug levels of phenytoin may be monitored, but such tests are less routinely available for other antiseizure medications.

Topiramate and zonisamide are associated with nephrolithiasis and should probably be avoided in patients with a history of or who are prone to this condition (see “Kidney stones in adults: Epidemiology and risk factors”). Renal tubular acidosis can also occur with these antiseizure medications; patients with preexisting conditions that make them prone to metabolic acidosis (eg, severe respiratory disorders, diarrhea) should also consider avoiding these drugs or have more frequent monitoring of serum bicarbonate levels [93].

In the setting of renal transplantation, potential drug interactions between antiseizure medications and immunosuppressive therapy should be considered. Enzyme-inducing antiseizure medications may lower serum immunosuppressant levels, while enzyme-inhibitors may increase levels.

Hepatic disease — Some antiseizure medications are associated with hepatic toxicity and should be avoided in patients with preexisting liver disease. These include valproate and felbamate, and to a lesser extent, phenytoin and carbamazepine [91,94]. Many other antiseizure medications are metabolized fully or partially in the liver (table 2), requiring caution and dose adjustment when used in patients with chronic liver disease. These include carbamazepine, lamotrigine, phenytoin, phenobarbitalclobazam, valproate, felbamate, zonisamidetopiramateoxcarbazepineeslicarbazepine, and brivaracetamLevetiracetamgabapentinpregabalin, and vigabatrin do not undergo hepatic metabolism and are less problematic for use in patients with chronic liver disease.

Psychiatric disorders — Persons with epilepsy have a higher than expected prevalence of comorbid psychiatric disorders [95-99]. The association may relate to shared perturbations in neurotransmitter action, alterations to neural networks or both [96,98,100]. In persons with epilepsy, the presence of depression correlates more strongly with a poor quality of life than the frequency of the seizures [101].

Some antiseizure medications (valproatelamotriginecarbamazepineoxcarbazepine) appear to have mood stabilizing properties [102-104]. Their efficacy in this regard is best established for bipolar disorder. However, many physicians view these medications as attractive in patients with comorbid anxiety and depression.

In contrast, some antiseizure medications, in particular those that potentiate gamma-aminobutyric acid (GABA) neurotransmission (phenobarbitaltiagabinevigabatrintopiramate), have been reported to cause or exacerbate a depressed mood and perhaps should be avoided in patients with comorbid depression [105]. Similarly, drugs that have been reported to provoke psychosis (levetiracetam, topiramate, vigabatrin, zonisamideethosuximide, and perampanel) may be less desirable in patients with that history.

Use of one of the antiseizure medications thought to be effective in mood stabilization does not substitute for a full psychiatric evaluation and independent treatment of a coexisting psychiatric disorder. Further impetus for this comes from the fact that as a class, antiseizure medications are associated with an increased risk of suicide. All patients with epilepsy treated with antiseizure medications should be monitored for changes in mood and suicidality. (See “Overview of the management of epilepsy in adults”, section on ‘Specific adverse reactions’.)

Drug interactions are also a potential concern in patients with psychiatric disorders. Enzyme-inducing antiseizure medications (table 2) can decrease the plasma concentration of many antidepressants including tricyclic agents and selective serotonin reuptake inhibitors, as well as antipsychotic drugs and benzodiazepines [64,65].

Migraine — Some studies suggest that migraine may be more prevalent in patients with epilepsy and vice versa [106,107]. Valproategabapentin, and topiramate are antiseizure medications that have demonstrated efficacy for migraine prevention in placebo-controlled trials (see “Preventive treatment of episodic migraine in adults”, section on ‘Anticonvulsants’). This may provide an opportunity to limit polypharmacy in individuals with both migraine and epilepsy.

Osteoporosis risk — Antiseizure medications in chronic use have been associated with bone loss. Initially this association was observed for enzyme-inducing antiseizure medications (table 2), but later was found to extend to valproate as well as to some of the newer nonenzyme-inducing antiseizure medications [108-110]. The evidence associating osteoporosis and antiseizure medication therapy may be strongest for phenytoin. Osteoporosis is particularly problematic for patients with epilepsy, as seizures are associated with falls and bone fractures [93,111,112].

While phenytoin should perhaps be avoided in patients in whom there is concern for bone loss, there are insufficient data to recommend avoiding or choosing any other specific antiseizure medication in order to limit the risk of osteoporosis [93]. Rather, monitoring of bone density, routine supplementation of calcium and vitamin D, and a consistent exercise regimen are suggested for all patients on chronic antiseizure medication therapy. (See “Antiseizure medications and bone disease”.)


●Diabetes – Because of its association with weight gain, insulin resistance, and polycystic ovarian syndrome, use of valproate in individuals with diabetes or obesity should be carefully considered [113]. Carbamazepinevigabatringabapentin, and pregabalin are also, but less frequently, associated with weight gain. (See ‘Side effect profiles’ above.)

Some antiseizure medications (gabapentinpregabalin, and possibly carbamazepine) have efficacy in treating pain associated with diabetic neuropathy. (See “Management of diabetic neuropathy”, section on ‘Pain management’.)

●Thyroid disease – While many antiseizure medications, in particular the enzyme-inducing agents, can alter thyroid hormone levels, this is generally subclinical and should not impact drug choice [113,114]. Enzyme-inducing agents should probably be avoided in patients with severe thyroid dysfunction.

●Cancer – The choice of antiseizure medication in patients being treated for systemic cancer is influenced by potential drug interactions between enzyme-inducing antiseizure medications (table 2) and chemotherapeutic agents that can lead to decreased efficacy of both treatments [89,90]. By inhibiting their metabolism, valproate may increase the toxicity of certain cancer chemotherapy agents. There also may be an increased potential for allergic cutaneous reactions when antiseizure medications are used during radiotherapy.

●HIV – Enzyme-inducing antiseizure medications and those that are highly protein-bound (table 2) may interact with antiretroviral therapy (ART) [115-117]. Of particular concern is that these drug interactions may cause minor reductions in the levels of protease inhibitors that could lead to loss of viral suppression and the emergence of drug resistance. There are also concerns that phenytoin-associated skin rash may be more common in HIV-positive patients. Lamotrigine doses may need to be increased with certain medications including ritonavir and atazanavir. While early in vitro studies suggested that valproate might increase viral replication, a series of patients treated with valproate maintained excellent control of both seizures and HIV [118].

●Cardiovascular disease – Clinicians should consider potential drug interactions between enzyme-inducing antiseizure medications and statins, calcium channel blockers, and warfarin [64,65,119]. While, carbamazepine has been associated with heart block and other bradyarrhythmias in susceptible individuals [120], clinically significant electrocardiography (ECG) changes are uncommon with carbamazepine in older adult patients who do not have a preexisting conduction defect [121].

Because the cytochrome P450 enzymes are involved in cholesterol synthesis, it is possible that enzyme-inducing antiseizure medications may thereby affect vascular risk. In one small series, switching patients from carbamazepine or phenytoin to noninducing antiseizure medications levetiracetam or lamotrigine was associated with improvements in serologic markers of vascular risk (eg, total cholesterol, triglycerides, C-reactive protein) [122]. Some studies have found that long-term monotherapy with carbamazepine, phenytoin, or valproate, is associated with markers of increased cardiovascular risk, such as carotid intimal thickening, abnormal cholesterol, homocysteine, and folate metabolism, and elevated levels of C-reactive protein [123,124]. However, no studies have clearly linked any specific antiseizure medications to a higher or lower risk of vascular events.

●Blood disorders – Certain antiseizure medications (carbamazepinephenytoinethosuximidevalproate) are associated with neutropenia and agranulocytosis, and should be avoided in patients with blood disorders [125,126]. (See “Drug-induced neutropenia and agranulocytosis”.)

Similarly, drugs associated with thrombocytopenia (eg, carbamazepinevalproatephenytoin) should be avoided in patients with a low platelet count or a history of other bleeding diatheses. (See “Drug-induced immune thrombocytopenia”.)

Women of childbearing age — A number of issues are important in women of childbearing age, especially if they are considering becoming or are already pregnant.

Folate should be prescribed to all women of childbearing age who are taking antiseizure medications. (See “Management of epilepsy during preconception, pregnancy, and the postpartum period”, section on ‘Folic acid supplementation’.)

Hormonal contraception — Women should be informed about the interactions between antiseizure medication therapies and hormonal pill, patch, or ring contraception and the availability of long-acting reversible contraception (LARC), which is highly effective and avoids most if not all drug-drug interactions, depending on the specific method. (See “Management of epilepsy during preconception, pregnancy, and the postpartum period”, section on ‘Preconception management’ and “Contraception: Counseling and selection”.)

The expected contraceptive failure rate of 0.7 per 100 woman-years using oral contraceptives is increased to 3.1 per 100 woman-years in patients who concomitantly take enzyme-inducing antiseizure medications (table 2) [127-130]. While vigabatrin is not an enzyme inducer, lower levels of ethinyl estradiol have been reported in volunteers taking this antiseizure medication [131] (see “Combined estrogen-progestin oral contraceptives: Patient selection, counseling, and use”, section on ‘Drug interactions’). If an enzyme-inducing antiseizure medication is nonetheless deemed to be the drug of choice in a woman taking combined hormonal pill, patch, or ring contraception, alternative regimens or forms of contraception should be considered. (See “Overview of the management of epilepsy in adults”, section on ‘Contraception’.)

In addition to the effect of antiseizure medications on hormonal contraceptive metabolism, combined hormonal contraceptives can increase the metabolism of lamotrigine, thereby reducing the plasma drug concentration. In other words, higher doses of lamotrigine may be needed in women taking combined estrogen-progesterone contraception, and continuous dosing may be preferable to avoid increased lamotrigine levels during pill-free intervals. (See “Antiseizure medications: Mechanism of action, pharmacology, and adverse effects”, section on ‘Lamotrigine’.)

Catamenial epilepsy — Many women with epilepsy report an association between the occurrence of their seizures and certain phases of their menstrual cycle [132,133]. Catamenial seizure clustering can occur in women with any seizure type and epilepsy syndrome but may be more common among women with focal compared with generalized epilepsy [134-139] and among those with left-sided temporal epilepsy compared with right-sided, multifocal, or extratemporal epilepsy [140,141].

In general, research suggests that catamenial seizure patterns result from cyclic changes in hormone levels during the menstrual cycle; changes in antiseizure medication levels due to endogenous metabolic effects may also contribute. Estrogen levels peak mid-cycle and then, in women who do not conceive, fall through the onset of menses. It is during the late part of the menstrual cycle (just before the onset of menses), during a relative drop in estrogen levels, that seizures most often cluster [134,142]. Periovulatory (mid-cycle) seizure clustering can also occur.

The mainstay of treatment of catamenial seizures is an antiseizure medication that is most effective for the woman’s epilepsy syndrome. However, when catamenial seizures are not controlled with antiseizure medications, clinicians may consider use of a continuous estrogen-progestin contraceptive on the theoretical basis that suppressing estrogen fluctuations will lead to better seizure control. The rationale and use of hormonal prophylaxis for catamenial epilepsy is similar to that in estrogen-associated migraine, which is reviewed separately. (See “Estrogen-associated migraine, including menstrual migraine”.)

Intermittent benzodiazepine treatment timed according to the vulnerable phase of the menstrual cycle is also a common strategy. Clobazam is the only benzodiazepine studied systematically for this purpose. In a double-blind cross-over study, clobazam (20 to 30 mg/day) was administered for 10 days in the high-risk phase of the menstrual cycle in 18 women with catamenial epilepsy [143]. Fourteen patients reported better seizure control with clobazam than placebo. Long-term follow-up of patients who continued to use this treatment strategy revealed seizure remission and/or significant reduction of seizures in five of nine patients [144]. These limited data support a fairly common practice of treating catamenial seizure exacerbations with intermittent benzodiazepines with a long-acting agent such as lorazepam. A reasonable dose of lorazepam in this setting is 0.5 to 1 mg two to three times daily.

Very limited data suggest that appropriately timed acetazolamide may have some benefit in catamenial epilepsy [145-147]. Although cyclic natural progesterone has been reported to reduce seizure frequency in observational studies, a randomized trial failed to confirm a benefit in 294 women with poorly controlled seizures [148]. Other investigational strategies include gonadotropin analogs and neurosteroids such as ganaxolone [135,149-151].

Pregnancy and postpartum — Treatment of epilepsy during pregnancy must balance competing risks. Seizures, particularly convulsive seizures, are believed to be harmful to the fetus. At the same time, both major and minor malformations are more common in fetuses exposed to antiseizure medications in utero compared with offspring of untreated women with epilepsy and women without epilepsy. The overall risk of major malformations is 4 to 6 percent in exposed infants; valproate is a major contributor to this risk. Polypharmacy increases the risk. The timing (early versus late in gestation) and dose of exposure are also likely to be important. While no antiseizure medication has been definitively shown to be safe in pregnancy, the evidence linking valproate to fetal malformations is sufficiently convincing to recommend avoiding its initiation and use in most women of childbearing potential [152]. (See “Risks associated with epilepsy during pregnancy and postpartum period”.)

The management of epilepsy in pregnancy and during breastfeeding is discussed separately. (See “Management of epilepsy during preconception, pregnancy, and the postpartum period”.)


Patient education — Successful treatment can be optimized by a systematic approach that includes patient education [153,154]. Before treatment is initiated, the physician needs to counsel the patient and family to increase their understanding of epilepsy and their ability to report necessary and relevant information. These discussions will improve the likelihood that the patient will comply with the plan of treatment.

The physician should impress upon the patient, family, and patient’s friends the critical need to follow the prescribed drug regimen. Nonadherence to antiseizure medication treatment regimen is associated with increased risk of mortality, as well as hospitalization and injury [155]. (See “Comorbidities and complications of epilepsy in adults”.)

Written instructions on how and when to take the drugs should be provided and should explain the dosing regimen and any potential adverse effects or drug-drug interactions. The patient must also be warned not to stop taking an antiseizure medication on their own initiative, and not to allow a prescription to run out or expire.

Patients should be urged not to start any other prescription, over-the-counter medications, dietary supplements, or herbal remedies without first contacting their physician because these might affect serum concentrations of their antiseizure medications [66,156]. (See ‘Pharmacokinetics’ above.)

Drug administration and dosing — Treatment should be started with a single drug (monotherapy). In general, the strategy is to gradually titrate the dosage to that which is maximally tolerated and/or produces optimal seizure control (start low and go slow). Pooled analysis from two large prospective studies found that with this approach, adverse event reporting was no higher in treated versus untreated patients [157]. Variables other than antiseizure medication treatment were found to be associated with adverse event reporting, most notably comorbid depression. The recommended initial dose and suggested titration schedule is presented separately. (See “Antiseizure medications: Mechanism of action, pharmacology, and adverse effects”.)

Seizure calendar — Patients and family members should be asked to record seizures and antiseizure medication doses on a calendar, which can then be brought or sent to the physician for review. Seizure triggers (eg, stress, sleep deprivation, alcohol, menses) should be indicated. The patient and family should note on the calendar the hour at which any symptoms occur.

The seizure calendar helps to monitor and encourage compliance, as well as identify triggers. The seizure calendar also may be used to track the patient’s response to drug therapy, including possible side effects. In one study of 71 patients completing daily seizure diaries, both lack of sleep and higher self-reported stress and anxiety were associated with seizure occurrence [158]. Seizures were also associated with the patients’ own prediction of the likelihood of seizure occurrence. Physicians should be aware that patients are often unaware of their seizures and may significantly underestimate the number of seizures that occur, especially those that occur during sleep or that disrupt consciousness [159].

Laboratory monitoring — A complete blood count, liver function tests, blood urea nitrogen (BUN), and measurement of creatinine and electrolytes levels should be done prior to starting antiseizure medication therapy. Albumin levels should also be obtained prior to starting treatment with one of the highly protein-bound antiseizure medications.

Regular follow-up visits should be scheduled to check drug concentrations, blood counts, and hepatic and renal function. These visits are also used to address concerns the patient may have about taking the medication and possible side effects, or psychosocial aspects of their disorder. Drug levels should be checked at least yearly in patients who are not having seizures and not undergoing medication dose changes. Chemistry and hematology studies are usually checked in association with drug levels.

Drug levels can be helpful in the management of antiseizure medications [160]:

●To establish an individual therapeutic concentration when a patient is in remission

●To assist in the diagnosis of clinical antiseizure medication toxicity

●To assess compliance

●To guide dose adjustments, particularly in the setting of drug formulation changes, when an interacting medication is added to or removed from a patient’s regimen, or during pregnancy

Generic substitutions — The use of generic medications as a treatment for people with epilepsy has attracted much attention and debate, and the evidence is mixed in terms of whether generic substitution of antiseizure medications has an adverse impact on seizure control and toxicity. Clinicians should consider the possibility of generic substitution as a cause of unexpected breakthrough seizures or toxicity, along with other possible explanations. In addition, clinicians may wish to obtain laboratory monitoring with plasma drug levels when a change is made in drug formulation. This topic is reviewed in more detail separately. (See “Overview of the management of epilepsy in adults”, section on ‘Generic substitution’.)

SOCIETY GUIDELINE LINKSLinks 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”.)

INFORMATION FOR PATIENTSUpToDate 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 5th to 6th 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 10th to 12th 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)”)

●Beyond the Basics topic (see “Patient education: Seizures in adults (Beyond the Basics)”)

SUMMARY AND RECOMMENDATIONSWhen the diagnosis of epilepsy is made, the initial treatment is a single antiseizure medication. When to start treatment, and with what agent, is a complex decision that is individualized in order to optimize both efficacy and tolerability.

●The decision of whether or not to start antiseizure medication therapy at the time of a first unprovoked seizure in an adult should be individualized. The decision is based on several factors: an assessment of the risk for recurrent seizure, the potential benefits of immediate antiseizure medication therapy in reducing the risk of recurrent seizure, the side effects of antiseizure medications, and patient preferences. (See ‘First-time unprovoked seizure’ above.)

•Antiseizure medication treatment is reasonable in patients after a single unprovoked seizure if they also have a potential symptomatic cause of epilepsy (eg, stroke or trauma history, brain tumor), epileptiform features on electroencephalogram, a relevant abnormality on neuroimaging study (computed tomography [CT] or magnetic resonance imaging [MRI]), or an abnormal neurologic examination. Many of these patients likely meet criteria for epilepsy according to the International League Against Epilepsy (ILAE) definition, which considers patients with a single unprovoked seizure and an estimated risk of recurrence ≥60 percent over ten years to have epilepsy, similar to those with two unprovoked seizures occurring >24 hours apart. (See ‘Risk of seizure recurrence’ above.)

•Antiseizure medication treatment after a single unprovoked seizure in patients may be deferred depending on the presence or absence of other risk factors and on individual patient preferences. (See ‘Benefit of early versus deferred treatment’ above.)

●We recommend initiating antiseizure medication therapy in individuals who have had two or more unprovoked seizures (Grade 1A). Such patients are at high risk for further unprovoked seizures. (See ‘Second unprovoked seizure’ above.)

●The selection of antiseizure medication considers the type of epilepsy or epilepsy syndrome (table 3) and potential side effects (table 4 and table 5), as well as other prescribed medications and comorbidities. Gender and patient age, and cost and availability of medication may also be relevant factors. (See ‘Drug-related considerations’ above and ‘Seizure-related considerations’ above.)

●In general, enzyme-inducing antiseizure medications (eg, phenytoincarbamazepinephenobarbitaloxcarbazepine) are the most problematic for interactions with drugs such as warfarin, hormonal contraception, anti-cancer drugs, and anti-infective drugs. Specific interactions of antiseizure medications with other medications may be determined using the Lexicomp drug interactions tool. (See ‘Pharmacokinetics’ above.)

●Because antiseizure medications are either metabolized by the liver or excreted by the kidneys, renal and hepatic disease impacts on both the choice of antiseizure medication as well as the prescribing regimen. (See ‘Renal disease’ above and ‘Hepatic disease’ above.)

●Women of childbearing age should be counseled regarding possible teratogenic effects of antiseizure medications. Folic acid supplementation is recommended for all women of child-bearing potential. Valproate should be avoided and only prescribed if no other antiseizure medication is effective for that particular patient. (See ‘Women of childbearing age’ above.)

●Regular outpatient follow-up appointments and the use of seizure calendars can help maximize the success of epilepsy treatment. (See ‘Initiation of antiseizure medication therapy’ above.)

●Patients with epilepsy have a higher than expected incidence of mood problems, anxiety, and depression. Antiseizure medications have been associated with suicidality. Patients treated with antiseizure medications should be monitored for changes in mood and suicidality. (See ‘Psychiatric disorders’ above.)


  1. Fisher RS, Acevedo C, Arzimanoglou A, et al. ILAE official report: a practical clinical definition of epilepsy. Epilepsia 2014; 55:475.
  2. Nevalainen O, Ansakorpi H, Simola M, et al. Epilepsy-related clinical characteristics and mortality: a systematic review and meta-analysis. Neurology 2014; 83:1968.
  3. Krumholz A, Wiebe S, Gronseth GS, et al. Evidence-based guideline: Management of an unprovoked first seizure in adults: Report of the Guideline Development Subcommittee of the American Academy of Neurology and the American Epilepsy Society. Neurology 2015; 84:1705.
  4. Marson A, Jacoby A, Johnson A, et al. Immediate versus deferred antiepileptic drug treatment for early epilepsy and single seizures: a randomised controlled trial. Lancet 2005; 365:2007.
  5. Randomized clinical trial on the efficacy of antiepileptic drugs in reducing the risk of relapse after a first unprovoked tonic-clonic seizure. First Seizure Trial Group (FIR.S.T. Group). Neurology 1993; 43:478.
  6. Kim LG, Johnson TL, Marson AG, et al. Prediction of risk of seizure recurrence after a single seizure and early epilepsy: further results from the MESS trial. Lancet Neurol 2006; 5:317.
  7. Berg AT. Risk of recurrence after a first unprovoked seizure. Epilepsia 2008; 49 Suppl 1:13.
  8. Hauser WA, Rich SS, Lee JR, et al. Risk of recurrent seizures after two unprovoked seizures. N Engl J Med 1998; 338:429.
  9. Hauser WA, Rich SS, Annegers JF, Anderson VE. Seizure recurrence after a 1st unprovoked seizure: an extended follow-up. Neurology 1990; 40:1163.
  10. Musicco M, Beghi E, Solari A, Viani F. Treatment of first tonic-clonic seizure does not improve the prognosis of epilepsy. First Seizure Trial Group (FIRST Group). Neurology 1997; 49:991.
  11. Ramos Lizana J, Cassinello Garciá E, Carrasco Marina LL, et al. Seizure recurrence after a first unprovoked seizure in childhood: a prospective study. Epilepsia 2000; 41:1005.
  12. Kho LK, Lawn ND, Dunne JW, Linto J. First seizure presentation: do multiple seizures within 24 hours predict recurrence? Neurology 2006; 67:1047.
  13. Leone MA, Solari A, Beghi E, FIRST Group. Treatment of the first tonic-clonic seizure does not affect long-term remission of epilepsy. Neurology 2006; 67:2227.
  14. Das CP, Sawhney IM, Lal V, Prabhakar S. Risk of recurrence of seizures following single unprovoked idiopathic seizure. Neurol India 2000; 48:357.
  15. Chandra B. First seizure in adults: to treat or not to treat. Clin Neurol Neurosurg 1992; 94 Suppl:S61.
  16. Gilad R, Lampl Y, Gabbay U, et al. Early treatment of a single generalized tonic-clonic seizure to prevent recurrence. Arch Neurol 1996; 53:1149.
  17. Leone MA, Giussani G, Nolan SJ, et al. Immediate antiepileptic drug treatment, versus placebo, deferred, or no treatment for first unprovoked seizure. Cochrane Database Syst Rev 2016; :CD007144.
  18. Leone MA, Vallalta R, Solari A, et al. Treatment of first tonic-clonic seizure does not affect mortality: long-term follow-up of a randomised clinical trial. J Neurol Neurosurg Psychiatry 2011; 82:924.
  19. Jacoby A, Gamble C, Doughty J, et al. Quality of life outcomes of immediate or delayed treatment of early epilepsy and single seizures. Neurology 2007; 68:1188.
  20. Hesdorffer DC, Benn EK, Cascino GD, Hauser WA. Is a first acute symptomatic seizure epilepsy? Mortality and risk for recurrent seizure. Epilepsia 2009; 50:1102.
  21. Fields MC, Labovitz DL, French JA. Hospital-onset seizures: an inpatient study. JAMA Neurol 2013; 70:360.
  22. Kwan P, Brodie MJ. Early identification of refractory epilepsy. N Engl J Med 2000; 342:314.
  23. Brodie MJ, Kwan P. Staged approach to epilepsy management. Neurology 2002; 58:S2.
  24. French JA, Kanner AM, Bautista J, et al. Efficacy and tolerability of the new antiepileptic drugs I: treatment of new onset epilepsy: report of the Therapeutics and Technology Assessment Subcommittee and Quality Standards Subcommittee of the American Academy of Neurology and the American Epilepsy Society. Neurology 2004; 62:1252.
  25. Glauser T, Ben-Menachem E, Bourgeois B, et al. Updated ILAE evidence review of antiepileptic drug efficacy and effectiveness as initial monotherapy for epileptic seizures and syndromes. Epilepsia 2013; 54:551.
  26. Kanner AM, Ashman E, Gloss D, et al. Practice guideline update summary: Efficacy and tolerability of the new antiepileptic drugs I: Treatment of new-onset epilepsy: Report of the Guideline Development, Dissemination, and Implementation Subcommittee of the American Academy of Neurology and the American Epilepsy Society. Neurology 2018; 91:74.
  27. Nolan SJ, Marson AG, Weston J, Tudur Smith C. Carbamazepine versus phenytoin monotherapy for epilepsy: an individual participant data review. Cochrane Database Syst Rev 2015; :CD001911.
  28. Nevitt SJ, Marson AG, Weston J, Tudur Smith C. Sodium valproate versus phenytoin monotherapy for epilepsy: an individual participant data review. Cochrane Database Syst Rev 2018; 8:CD001769.
  29. Chadwick DW, Anhut H, Greiner MJ, et al. A double-blind trial of gabapentin monotherapy for newly diagnosed partial seizures. International Gabapentin Monotherapy Study Group 945-77. Neurology 1998; 51:1282.
  30. French J, Glue P, Friedman D, et al. Adjunctive pregabalin vs gabapentin for focal seizures: Interpretation of comparative outcomes. Neurology 2016; 87:1242.
  31. Brodie MJ, Overstall PW, Giorgi L. Multicentre, double-blind, randomised comparison between lamotrigine and carbamazepine in elderly patients with newly diagnosed epilepsy. The UK Lamotrigine Elderly Study Group. Epilepsy Res 1999; 37:81.
  32. Brodie MJ, Richens A, Yuen AW. Double-blind comparison of lamotrigine and carbamazepine in newly diagnosed epilepsy. UK Lamotrigine/Carbamazepine Monotherapy Trial Group. Lancet 1995; 345:476.
  33. Steiner TJ, Dellaportas CI, Findley LJ, et al. Lamotrigine monotherapy in newly diagnosed untreated epilepsy: a double-blind comparison with phenytoin. Epilepsia 1999; 40:601.
  34. Brodie MJ, Chadwick DW, Anhut H, et al. Gabapentin versus lamotrigine monotherapy: a double-blind comparison in newly diagnosed epilepsy. Epilepsia 2002; 43:993.
  35. Kwan P, Brodie MJ, Kälviäinen R, et al. Efficacy and safety of pregabalin versus lamotrigine in patients with newly diagnosed partial seizures: a phase 3, double-blind, randomised, parallel-group trial. Lancet Neurol 2011; 10:881.
  36. Privitera MD, Brodie MJ, Mattson RH, et al. Topiramate, carbamazepine and valproate monotherapy: double-blind comparison in newly diagnosed epilepsy. Acta Neurol Scand 2003; 107:165.
  37. Ramsay E, Faught E, Krumholz A, et al. Efficacy, tolerability, and safety of rapid initiation of topiramate versus phenytoin in patients with new-onset epilepsy: a randomized double-blind clinical trial. Epilepsia 2010; 51:1970.
  38. Bill PA, Vigonius U, Pohlmann H, et al. A double-blind controlled clinical trial of oxcarbazepine versus phenytoin in adults with previously untreated epilepsy. Epilepsy Res 1997; 27:195.
  39. Guerreiro MM, Vigonius U, Pohlmann H, et al. A double-blind controlled clinical trial of oxcarbazepine versus phenytoin in children and adolescents with epilepsy. Epilepsy Res 1997; 27:205.
  40. Christe W, Krämer G, Vigonius U, et al. A double-blind controlled clinical trial: oxcarbazepine versus sodium valproate in adults with newly diagnosed epilepsy. Epilepsy Res 1997; 26:451.
  41. Dam M, Ekberg R, Løyning Y, et al. A double-blind study comparing oxcarbazepine and carbamazepine in patients with newly diagnosed, previously untreated epilepsy. Epilepsy Res 1989; 3:70.
  42. Baulac M, Brodie MJ, Patten A, et al. Efficacy and tolerability of zonisamide versus controlled-release carbamazepine for newly diagnosed partial epilepsy: a phase 3, randomised, double-blind, non-inferiority trial. Lancet Neurol 2012; 11:579.
  43. Baulac M, Patten A, Giorgi L. Long-term safety and efficacy of zonisamide versus carbamazepine monotherapy for treatment of partial seizures in adults with newly diagnosed epilepsy: results of a phase III, randomized, double-blind study. Epilepsia 2014; 55:1534.
  44. Brodie MJ, Perucca E, Ryvlin P, et al. Comparison of levetiracetam and controlled-release carbamazepine in newly diagnosed epilepsy. Neurology 2007; 68:402.
  45. Trinka E, Marson AG, Van Paesschen W, et al. KOMET: an unblinded, randomised, two parallel-group, stratified trial comparing the effectiveness of levetiracetam with controlled-release carbamazepine and extended-release sodium valproate as monotherapy in patients with newly diagnosed epilepsy. J Neurol Neurosurg Psychiatry 2013; 84:1138.
  46. Brodie MJ, Wroe SJ, Dean AD, et al. Efficacy and Safety of Remacemide versus Carbamazepine in Newly Diagnosed Epilepsy: Comparison by Sequential Analysis. Epilepsy Behav 2002; 3:140.
  47. Baulac M, Rosenow F, Toledo M, et al. Efficacy, safety, and tolerability of lacosamide monotherapy versus controlled-release carbamazepine in patients with newly diagnosed epilepsy: a phase 3, randomised, double-blind, non-inferiority trial. Lancet Neurol 2017; 16:43.
  48. Lattanzi S, Zaccara G, Giovannelli F, et al. Antiepileptic monotherapy in newly diagnosed focal epilepsy. A network meta-analysis. Acta Neurol Scand 2019; 139:33.
  49. Perucca E, Brodie MJ, Kwan P, Tomson T. 30 years of second-generation antiseizure medications: impact and future perspectives. Lancet Neurol 2020; 19:544.
  50. Privitera MD. Evidence-based medicine and antiepileptic drugs. Epilepsia 1999; 40 Suppl 5:S47.
  51. Williamson PR, Marson AG, Tudur C, et al. Individual patient data meta-analysis of randomized anti-epileptic drug monotherapy trials. J Eval Clin Pract 2000; 6:205.
  52. Gamble C, Williamson PR, Chadwick DW, Marson AG. A meta-analysis of individual patient responses to lamotrigine or carbamazepine monotherapy. Neurology 2006; 66:1310.
  53. Muller M, Marson AG, Williamson PR. Oxcarbazepine versus phenytoin monotherapy for epilepsy. Cochrane Database Syst Rev 2006; :CD003615.
  54. Tudur Smith C, Marson AG, Clough HE, Williamson PR. Carbamazepine versus phenytoin monotherapy for epilepsy. Cochrane Database Syst Rev 2002; :CD001911.
  55. Marson AG, Appleton R, Baker GA, et al. A randomised controlled trial examining the longer-term outcomes of standard versus new antiepileptic drugs. The SANAD trial. Health Technol Assess 2007; 11:iii.
  56. Marson AG, Al-Kharusi AM, Alwaidh M, et al. The SANAD study of effectiveness of carbamazepine, gabapentin, lamotrigine, oxcarbazepine, or topiramate for treatment of partial epilepsy: an unblinded randomised controlled trial. Lancet 2007; 369:1000.
  57. Marson AG, Al-Kharusi AM, Alwaidh M, et al. The SANAD study of effectiveness of valproate, lamotrigine, or topiramate for generalised and unclassifiable epilepsy: an unblinded randomised controlled trial. Lancet 2007; 369:1016.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: