Antiepileptics

Epilepsy and seizures are more common in older adults than in younger individuals [584]. Some of the most common risk factors for new-onset epilepsy in older people include neurological conditions such as cerebrovascular disease (e.g. stroke) and dementia [584]. Antiepileptic medicines are effective in managing epilepsy for approximately two-thirds of individuals [585], but they can also cause significant adverse effects [586]. Studies reported that up to 90% of people taking antiepileptics experience adverse effects, including dizziness, sedation, cognitive impairment, and neuropsychiatric symptoms [586, 587]. The primary goal of treatment is to induce remission while minimising side effects.

There is insufficient evidence to support their routine prophylactic use of seizures in conditions such as traumatic brain injury, subarachnoid haemorrhage, brain tumours, and post-stroke [588]. A study on valproic acid found that using it for more than three months after a stroke did not significantly reduce the risk of late-onset seizures [589]. Importantly, individuals who discontinued valproic acid within three months did not have a higher risk of late seizures compared to those who continued treatment beyond three months [589]

Apart from epilepsy, certain antiepileptics are prescribed for other indications including neuropathic pain, bipolar disorder, or migraine prevention. In the context of neuropathic pain, a 2017 randomised, double-blind, placebo-controlled trial of pregabalin found no significant difference in leg pain severity associated with sciatica between the pregabalin and placebo groups [590]. Given the potential for side effects, deprescribing should be considered when antiepileptics provide little or no apparent benefit, particularly if individuals experience adverse effects such as dizziness.

Non-epilepsy indications

Refer to the narrative evidence summary, the GRADE Summary of Findings table in the guidelines, and the Technical Report for a complete presentation of the deprescribing evidence based on the GRADE framework (including other factors considered in developing the recommendations).

When there is clear risk-benefit balance for the use of antiepileptics, it may be appropriate to continue therapy with regular monitoring in place. For instance, primidone is also indicated for the management of essential tremor [591].

Epilepsy indication

When seizures have been well-controlled for several years, some individuals may wish to discontinue antiepileptic medications. However, this decision requires a thorough discussion of the associated risks and benefits. Discontinuation is a complex process that should involve consultation with a neurologist, who can assess factors that increase the likelihood of successful withdrawal. These factors include monotherapy, the absence of epileptiform abnormalities on electroencephalogram prior to withdrawal, a long seizure-free period, and no history of focal seizures [586, 592]. Additionally, developmental delay, the duration of epilepsy before remission, age at onset of seizures, and a history of febrile seizures also play crucial roles in the decision-making process [593].

The potential benefits of discontinuation should be weighed against the risk of seizure recurrence, which ranges from 30% to 50%, depending on the individual's condition and risk factors [588]. For those who have been seizure-free for at least two years, deprescribing may be considered, particularly in those at low risk [594]. Each additional year of seizure freedom further reduces the likelihood of seizure recurrence [593, 595], suggesting that a longer seizure-free period is preferable before initiating withdrawal. However, this must be balanced against the long-term adverse effects of continued antiepileptic use. While the two-year threshold is commonly referenced, it is an arbitrary threshold; the risk of recurrence continues to decrease with each additional seizure-free year [593]. Lamberink et al. developed nomograms based on a systematic review and individual participant data meta-analysis, which can assist in determining the optimal timing for deprescribing in individual patients [593].

Apart from the risk of seizure recurrence, the decision to stop antiepileptics should take into account the type of epilepsy, seizure history as well as personal and social factors important to the person (e.g. relationships, driving, and employment) given potentially devastating consequences of recurrent seizures [596]. In Australia, under the driver licensing authority standards for driving a private vehicle, a person should not drive a vehicle while their antiepileptic dose is being tapered and for three months after the final dose [597]. However, there are two exceptions to this restriction: 1) the dose reduction is due only to the presence of current dose-related side effects and is unlikely to affect seizure control; or 2) the dose reduction is required, after an increase due to a temporary situation, to a dose that was effective before the increase [597]. The risks and implications of seizure recurrence versus the benefits of discontinuing therapy should be communicated to the person to ensure informed consent and shared decision-making. If deprescribing is considered appropriate in the context of epilepsy, gradual dose reduction should be individual with close monitoring for breakthrough seizures. The speed of tapering should be guided by the person by taking into consideration other factors such as driving implications. A slow tapering (such as by 10-25% monthly over several months) allows for monitoring and identifying the minimal effective dose in case of seizure recurrence, but it prolongs the non-driving period for the person may have significant personal impacts [586].

Drug interactions should also be considered, as some antiepileptics, especially older agents like phenytoin and carbamazepine, can alter the metabolism of other medicines through enzyme induction [598].

We identified one RCT related to carbamazepine deprescribing from the systematic review and meta-analysis [599] and one before-and-after study related to gabapentinoid deprescribing [600]; however, no evidence related to the withdrawal of antiepileptics in people treated for epilepsy.

There is currently insufficient evidence to guide deprescribing of antiepileptics in people with epilepsy. Existing data come from one RCT and one before-and-after study in older adults using antiepileptics for non-epilepsy indications (e.g. behavioural and psychological symptoms of dementia, neuropathic pain). Both studies are of very low certainty due to small sample sizes and short follow-up. While these studies suggest that deprescribing may be feasible and not associated with harm in these non-epilepsy contexts, the evidence is too limited to support formal recommendations. If deprescribing is considered appropriate, close monitoring of symptom recurrence should be undertaken as well as a review of concurrent medicines that require adjustments due to changes in drug-drug interactions when tapering antiepileptics.

For non-epilepsy indications, the dose of antiepileptics may be gradually reduced (e.g. by 25% of the previous dose every week) as it may be helpful to determine the minimal effective dose in case of symptom recurrence. In people who have been taking antiepileptics for longer than four weeks, especially for barbiturates and benzodiazepines, it may be reasonable to consider tapering more gradually over several months to minimise the risk of adverse drug withdrawal events [601].