The Anti-Seizure Drugs Compared

Whatever drug your vet reaches for, the strategy is usually the same, and it explains a lot of what follows. The standard approach is monotherapy first: start a single first-line drug at an adequate dose, allow it time to build up to a steady level in the blood, monitor the response, and adjust to effect before anyone reaches for a second drug. Add-on drugs come in only when a single agent, at a proper dose and level, has failed to control the seizures (Bhatti et al., 2015). So if your vet starts with one drug and seems in no hurry to pile on more, that patience is the textbook approach, not a lack of urgency.

The second principle I'll keep repeating, because it genuinely matters: never stop an anti-seizure drug abruptly. Stopping suddenly, or letting doses lapse, can trigger withdrawal seizures, a cluster, or even status epilepticus, the dangerous prolonged seizure state. If a drug ever needs to be withdrawn or swapped, it's done by slow tapering, with the consensus advising the dose come down by around 20% or less per month (Bhatti et al., 2015). This holds no matter how well your pet seems, and no matter how much you dislike a side effect. The answer to a side effect is a phone call to your vet, never a sudden stop.

And the choice is individualised. There's no league table with a single winner at the top. The right drug depends on the species, the seizure pattern, other health problems, and a few practical realities I'll come to. So treat what follows as a fair description of the contenders, not a recommendation.

For dogs, the menu of licensed first-line drugs in the UK and Europe is short. There are only two: phenobarbital and imepitoin, the latter sold as Pexion (Bhatti et al., 2015). Everything else is an add-on or used off the licence, so the first real decision usually comes down to a straight choice between these two.

Phenobarbital (phenobarbitone) is the long-established mainstay, for good reason. When the blood level is kept in the right range, it reduces seizure frequency in around 60 to 93% of dogs with idiopathic epilepsy (Bhatti et al., 2015). The catch is that it needs therapeutic drug monitoring: blood tests to check the level sits in the target range, roughly 25 to 30 mg/l for most dogs, with levels above 35 mg/l carrying an increased risk of liver damage and best avoided (Bhatti et al., 2015). Early on it can cause sedation, wobbliness (ataxia), a bigger appetite, and increased thirst and urination, though many of these settle as the dog adjusts (Bhatti et al., 2015; Tipold et al., 2015). Because it's processed by the liver, liver monitoring is part of the deal. That liver question deserves more space than a fair comparison should give it, so I'm handing the full picture to phenobarbital, what to expect and the blood-testing detail to therapeutic drug monitoring.

Imepitoin (Pexion) is the newer option, developed and licensed specifically for dogs (Bhatti et al., 2015; Charalambous et al., 2018). It's licensed first-line for dogs with recurrent single generalised seizures, and the evidence behind it is good. In the pivotal blinded, randomised trial of 226 dogs, imepitoin came out non-inferior to phenobarbital, meaning comparable seizure control (Tipold et al., 2015). Its appeal is tolerability and convenience. It's generally better tolerated, and it needs no routine blood-level monitoring, because it has a wide safety margin and its blood concentration doesn't track neatly with its effect (Tipold et al., 2015; Bhatti et al., 2015). In that head-to-head trial, sedation, increased thirst and increased appetite were all significantly more common in the phenobarbital group, and it was phenobarbital, not imepitoin, that pushed up the liver enzymes (Tipold et al., 2015).

So why isn't imepitoin simply the obvious choice for everyone? One important limit. Its effectiveness hasn't been shown for cluster seizures or status epilepticus. The consensus is careful with its wording: "its efficacy has not yet been demonstrated in dogs with cluster seizures or status epilepticus" (Bhatti et al., 2015). For a dog whose seizures come in clusters, that's a real mark against it. It's given twice daily, usually starting around 10 mg/kg and titrated up towards 30 mg/kg, and it shouldn't be used in dogs with severe liver, kidney or heart problems (Boehringer Ingelheim, n.d.; Bhatti et al., 2015). The full imepitoin story, a dog-only one, lives in imepitoin (Pexion) explained.

When a first-line drug at a proper dose isn't enough on its own, two other drugs come into play. Both are worth knowing about even now, because they shape where treatment can go.

Potassium bromide (KBr, sold in the UK as Libromide) is the old guard. In most of the UK and Europe it's licensed only as an add-on, for dogs whose epilepsy hasn't responded well to first-line treatment, most often on top of phenobarbital (Bhatti et al., 2015; Dechra, n.d.). It has one standout virtue: it isn't processed by the liver at all but excreted unchanged by the kidneys, which makes it useful for a dog whose liver is already struggling (Bhatti et al., 2015). The trade-off is that it's fussy about diet. Bromide is reabsorbed in the kidney in competition with chloride (salt), so a change in how much salt your dog eats shifts the blood level. Dogs on bromide need a consistent diet to keep the level stable, and it needs blood-level monitoring as phenobarbital does (Bhatti et al., 2015). I'll come to the one absolute rule about bromide and cats shortly, because it matters more than anything else on this page.

Levetiracetam (Keppra) is the well-tolerated all-rounder. Side effects are uncommon and usually mild: some sedation, a touch of wobbliness, a slightly reduced appetite, the odd bout of vomiting (Bhatti et al., 2015). It's used as an add-on, with the consensus judging there to be "fair" evidence to recommend it as an adjunctive drug (Bhatti et al., 2015). In the UK it isn't licensed for veterinary use, so it's prescribed under the cascade, the legal route vets use to reach a human-licensed medicine when no suitable animal-licensed one fits (Bhatti et al., 2015; VMD, n.d.). Two honest caveats. The first is the "honeymoon effect": some dogs respond well at first, then gradually develop tolerance over months, so the benefit can fade (Bhatti et al., 2015). The second is practical: its short action means it's usually given three times a day rather than twice (Bhatti et al., 2015). On the plus side, it lends itself to pulse dosing for dogs that cluster, an initial 60 mg/kg once a seizure starts or warning signs appear, then 20 mg/kg three times daily until the dog has been seizure-free for 48 hours (Bhatti et al., 2015). The at-home rescue side of this belongs in managing cluster seizures at home; I mention it here only as a comparison point.

Here's the same picture laid out plainly.

Drug	Species	Role	Monitoring	Key notes
Phenobarbital	Dog and cat	First-line	Blood levels needed (target ~25 to 30 mg/l; avoid above 35)	Effective in roughly 60 to 93% of dogs; processed by the liver; early sedation, appetite and thirst that often settle; feline first-line (Bhatti et al., 2015; Charalambous et al., 2018).
Imepitoin (Pexion)	Dog only	First-line	No routine monitoring	Better tolerated, comparable to phenobarbital; not shown to work for clusters or status; not licensed in cats (Tipold et al., 2015; Bhatti et al., 2015).
Potassium bromide (Libromide)	Dog only; contraindicated in cats	Add-on (UK)	Blood levels needed	Kidney-excreted so spares the liver; sensitive to dietary salt; causes serious, sometimes fatal airway disease in cats (Bhatti et al., 2015; Bertolani et al., 2012).
Levetiracetam (Keppra)	Dog and cat	Add-on / pulse	No levels needed	Well tolerated; honeymoon (tolerance) effect; useful for clusters; three times daily; UK cascade/off-label; feline second-line (Bhatti et al., 2015; Charalambous et al., 2018).

If you have a cat, please read this section carefully, because feline epilepsy is not simply dog epilepsy in a smaller patient. The drug choices differ, and one of the dog drugs is genuinely dangerous to cats. The cat anchor what causes seizures in cats goes deeper, but the safety essentials belong right here.

For cats, phenobarbital is the first-line choice, and it's generally well tolerated. In the feline systematic review, the majority of cats across the studies were treated successfully on it (Charalambous et al., 2018). Levetiracetam is the main second-line or add-on drug: safe, with few side effects, usable on its own or alongside phenobarbital, and particularly valuable for the myoclonic and audiogenic reflex seizures sometimes seen in older cats (Charalambous et al., 2018).

Now the rule. Potassium bromide is contraindicated in cats. It causes a serious and sometimes fatal lower-airway disease, with coughing, laboured breathing and inflammation deep in the lungs, and it appears to be a hypersensitivity reaction with no safe dose. In a retrospective series of seven cats on bromide, every one developed coughing; the signs improved when the drug was stopped, but one cat went on to develop lipid pneumonia, suffered a collapsed lung and was put to sleep (Bertolani et al., 2012). Earlier series had already reported respiratory disease in a substantial minority of bromide-treated cats, with deaths among them (Bertolani et al., 2012). The authors' conclusion is unambiguous: "using bromide for seizure control in cats cannot be recommended" (Bertolani et al., 2012). Don't let anyone present bromide as a feline option.

The same caution applies to imepitoin (Pexion), which is not licensed for cats. It was developed for dogs, and the feline evidence amounts to a single small trial in eight cats, nowhere near enough to make it a routine choice (Charalambous et al., 2018). So neither bromide nor imepitoin belongs on a cat's prescription.

One more feline danger isn't an epilepsy drug at all, but it causes so many emergency seizures in cats that it has to be flagged here. Permethrin and other pyrethroids, the insecticides in many dog "spot-on" flea treatments, are toxic to cats. A product meant for a dog, applied to a cat or groomed off a dog by a cat in the same house, can cause violent tremors and seizures (Vettimes, n.d.). Cats can't break these chemicals down efficiently, so what's safe for a dog is dangerous for them. This is a reactive seizure from a poison, not epilepsy, and it's a genuine emergency, but it's entirely preventable: never use a dog flea product on a cat. The detail lives in what causes seizures in cats.

So how does a vet land on one drug rather than another? Once you see the reasoning, the choice stops feeling arbitrary. Several threads get woven together.

Species comes first. As you've just seen, dog and cat have different menus, and getting that right is non-negotiable.

Then the seizure pattern. A dog whose seizures come in clusters, or who has had status, points away from imepitoin, whose effect on clusters and status hasn't been shown, and towards phenobarbital, often with levetiracetam or bromide alongside (Bhatti et al., 2015). A dog with isolated, single seizures is a fair candidate for either first-line drug. This is exactly why the record you keep matters so much.

Then other health problems. Existing liver disease tips the balance away from phenobarbital and towards bromide or imepitoin, since bromide is cleared by the kidneys and spares the liver (Bhatti et al., 2015). Severe kidney disease, by contrast, is a reason to be cautious with bromide (Bhatti et al., 2015).

Then the practical realities. Phenobarbital and bromide need regular blood-level monitoring; imepitoin doesn't (Bhatti et al., 2015). Phenobarbital and imepitoin are given twice a day; levetiracetam usually three times (Bhatti et al., 2015). And there's cost and licensing: the licensed options (phenobarbital, Pexion, and Libromide as an add-on) sit alongside cascade or off-label use for levetiracetam (VMD, n.d.; Dechra, n.d.). The lifetime financial picture is its own subject, covered in the cost of epilepsy and how insurance works.

And underneath all of it, the strength of the evidence, honestly stated. The systematic-review evidence is good for phenobarbital and imepitoin, fair for potassium bromide and levetiracetam, and insufficient for most other drugs that occasionally get mentioned (Charalambous et al., 2014). That's why the real choice, for most animals, sits among these four.

Here's the thread that ties the whole decision together, and it's the one piece genuinely in your hands. Every driver above, the seizure pattern, whether a drug is working, whether you've reached that 50% responder mark, comes back to one thing: an accurate record of your pet's seizures. Frequency is the number your vet doses against and the yardstick for judging whether the chosen drug is doing its job. Guesswork and memory aren't enough, especially when you're frightened and exhausted and the months blur together.

That's what the Seizure Diary is built for. Logging each seizure, its date and time, how long it lasted, how your pet recovered, and which medication was given or missed, turns a vague impression into the clear picture your vet needs to choose well and adjust sensibly. It's the difference between "I think he's a bit better" and "his seizures have halved since the dose went up." Start it the day treatment begins.

Whichever drug you and your vet settle on, hold the two honest truths from the top of this page. The aim is fewer, milder seizures, not a cure, and that's still a very good outcome. And you never stop the drug suddenly; you pick up the phone instead. From here, the natural next steps are the deep dives on whichever drug is yours, phenobarbital or imepitoin, and, if you're still deciding whether to treat at all, when should treatment start. The choice in front of you has a logic to it, and now you can see it.