Choosing whether to prescribe at all

Antidepressants are not contra-indicated in stable epilepsy; however, the BNF advises caution when using antidepressants in people with epilepsy or a history of seizures, with very few exceptions.

The relatively low risk of antidepressants affecting seizure threshold rarely outweigh the risk of leaving depression untreated.  Antidepressants can be used in people with epilepsy, although NICE suggest specialist advice (e.g. neurologist) should be sought for those taking Anti-epileptic drugs (AEDs).

Refer to NICE Clinical Knowledge Summary for management of depression in general. This page assumes the decision to prescribe antidepressants has already been made.

Choosing between antidepressants

There is no clear consensus across clinical guidelines on the best choice of antidepressant in people with epilepsy.

A Cochrane review has concluded that there is no high quality evidence to guide on the best choice of antidepressant in people with epilepsy. The Scottish Intercollegiate Guidelines network (SIGN) reached a similar conclusion although they suggest that in general antidepressants are safe in people with epilepsy.

Low to moderate risk antidepressants in epilepsy

These antidepressants are considered suitable options in epilepsy by a range of mental health resources including Bazire’s Psychotropic Drug Directory, Maudsley Prescribing Guidelines in Psychiatry and published papers.

Selective Serotonin Inhibitors (SSRIs)

Preferred options, in no order preference, are:

• Citalopram
• Escitalopram
• Fluoxetine
• Sertraline

Scottish Intercollegiate Guidelines network (SIGN) suggest that SSRIs appear to be safe to use in people with epilepsy and depression. The NICE CKS guidance on depression recommends SSRIs, such as sertraline or citalopram, as being suitable for people with chronic health problems.

Other antidepressants

Other ‘low to moderate risk’ antidepressants, in no order of preference for use in epilepsy, are as below:   

• Selective and Noradrenaline Reuptake Inhibitors (duloxetine is preferred over venlafaxine)
• Mirtazapine
• Reboxetine
• Vortioxetine
• Agomelatine
• Doxepin  (preferred option of the tricyclic antidepressant group)
• Monoamine Oxidase Inhibitors (moclobemide preferred over phenelzine, isocarboxazid and trancylcypromine, which are seldom used in practice due to the risk of interactions with food and drink)

High risk antidepressants to avoid in epilepsy

Tricyclic antidepressants (particularly amitriptyline and clomipramine) should be avoided as they lower the seizure threshold and are deemed the most pro-convulsive.

Other considerations when making a choice

Drug interactions

• Check for potential drug interactions between current AEDs and the chosen antidepressant prior to initiation. Use drug interaction resources such as The BNF Interaction checker to do this.
• Complex drug interactions are possible; for example:
  • Some antidepressants can increase AED blood levels, affecting drugs with a narrow therapeutic range (e.g. carbamazepine, phenytoin, valproate).
  • Some enzyme inducing AEDs can lower antidepressant blood levels, possibly leading to treatment failure.

Avoid using multiple antidepressants

• The risk of seizure increases with multiple concurrent antidepressants.

Consider the need for neurologist input

• Where the patient’s AED has adverse psychiatric side-effects associated with depression (e.g. levetiracetam, phenytoin, phenobarbitone, primidone, topiramate, and vigabatrin).
• Where the patient might benefit from an AED with mood stabilising properties (e.g. carbamazepine, gabapentin, lamotrigine, oxcarbazepine, valproate).

Initiating the antidepressant

Use a low dose first

• Initiate the chosen antidepressant at a low dose and increase gradually until a standard therapeutic dose is achieved. Review regularly thereafter as recommended by NICE CKS guidance on depression.
• The recommendations on choices above assume use of antidepressants at standard doses.

Use the lowest therapeutic dose possible

• There is a dose dependent relationship between antidepressants and seizures. The British National Formulary (BNF) contains information on licensed dosing.

Monitoring epilepsy and AEDs

Monitoring seizure frequency

• Document the baseline
• Ask the patient to keep a seizure diary

If seizures occur or seizure incidence increases

• Consider checking sodium levels for hyponatraemia. Antidepressants (often SSRIs) can cause hyponatraemia and seizures may occur where this is severe.
• Consider the need for neurologist input.
• Consider switching the antidepressant.

Monitoring blood levels of AEDs

• For AEDs with narrow therapeutic range (e.g. carbamazepine, phenytoin), consider blood monitoring, particularly if there are concerns of potential toxicity (e.g. the risk of an interaction with a newly started antidepressant).
• If dosage adjustment of the AED may be required, seek advice from the neurologist.

Switching between antidepressants

Where people with stable epilepsy do not respond to their initial antidepressant, a switch in antidepressant may be necessary. Guidance on switching antidepressants specifically in epilepsy does not exist, but it may be helpful to:

Refer to NICE CKS Guidance

• Refer to NICE CKS guidance on altering antidepressant in therapy in the general population

Refer to our drug switching content

• Switching strategies for antidepressants
• MAOI to other antidepressants: switching in adults
• Moclobemide to other antidepressants: switching in adults
• Trazodone to other antidepressants: switching in adults
• Vortioxetine to other antidepressants: switching in adults
• SNRIs to other antidepressants: switching in adults
• Agomelatine to other antidepressants: switching in adults
• Mirtazapine to other antidepressants: switching in adults
• Considerations when choosing an alternative antidepressant
• Establishing if a person needs to switch their antidepressant

Consult mental health specialists

• Complex scenarios of treatment failure may require specialist input.

Further reading

The advice above has been generated using a range of literature. We’ve provided a brief summary to support our review below together with a full bibliography:

• Studies looking specifically at using antidepressants in people with epilepsy are limited. They are usually small, of low quality, and of retrospective design with varying definitions and degrees of depression, making comparison difficult.
  • As such, some clinical data exist for these antidepressants in epilepsy: citalopram, fluoxetine, mirtazapine, sertraline.
  • Of note, there are lack of clinical data in epilepsy for agomelatine, escitalopram, fluvoxamine, doxepin, duloxetine, paroxetine, MAOIs, venlafaxine, reboxetine, vortioxetine.
• A recent systematic review with broad study inclusion criteria (RCTs, observational studies and case reports in people with and without epilepsy) sought to determine the seizure risk associated with antidepressants. The review concluded that the risk of seizures is generally low for antidepressants used at therapeutic doses, but the risk is not zero for any of them.
  • For fluoxetine and duloxetine the risk of seizure was deemed almost negligible
  • For escitalopram, paroxetine, mirtazapine, sertraline and citalopram, a ‘low to moderate’ risk was assigned.
• There is generally a lack of published studies on antidepressant use in people with epilepsy; the relationship between antidepressants and seizure activity is often based on research in the general population as opposed to in people with epilepsy. A 5 year primary care cohort study of people with depression and no history of epilepsy or seizures (n=235,489) was conducted in the UK. At 5 years, 0.37% of those prescribed antidepressants had new seizures. All antidepressants except sertraline, escitalopram and mirtazapine were significantly associated with risk of epilepsy/seizures (compared to not using them). Trazodone, lofepramine and venlafaxine were associated with highest risk.

Acknowledgements

Thanks to Professor David Taylor, Director of Pharmacy at the Maudsley Hospital

Update history