Companion animal practice

Treatment and monitoring of

epilepsy in dogs

Kate Chandler

Epilepsy is a brain disorder and is defined as the propensity to have recurrent

seizures. It is the most common chronic neurological disorder seen in dogs.
Most dogs that present with recurrent seizures have idiopathic epilepsy,
Kate Chandler graduated from which is thought to have a genetic basis and has no identifiable underlying
the Royal Veterinary College (RVC) cause when a full diagnostic work-up is undertaken. Symptomatic epilepsies,
in 1995. She holds the European which arise secondary to brain diseases such as intra­cranial neoplastic
diploma in veterinary neurology lesions or central nervous system inflammatory disorders, are less common.
and a PhD for studies By addressing some of the most frequently asked questions relating to the
on GABA receptor function treatment of seizures, this article suggests some practical and effective
in status epilepticus. She was strategies for managing and monitoring dogs with idiopathic epilepsy.
a lecturer in neurology and
neurosurgery at the RVC for
six years, where she ran canine
electroencephalography and Types of seizure
epilepsy clinics. She recently moved
to work for the UK government. Seizures may be focal or generalised. Focal seizures
arise when only part of one cerebral hemisphere devel-
ops abnormal epileptiform activity, while generalised
seizures arise when there is significant abnormal acti-
vation of both cerebral hemispheres. There are sev-
eral different types of focal and generalised seizure
(Chandler 2006). Typically, generalised seizures have
tonic and clonic components (periods of rigidity and
rhythmic movements, respectively), are symmetrical
and involve most of the body. Focal seizures are often
asymmetrical and may appear as limited movements
of isolated parts of the body, such as lip twitching or
rhythmic movements of a limb.

Should treatment be initiated?

Fig 1: Dog with suspected epilepsy undergoing
The key point when considering initiating treatment
an electroencephalographic examination
for epilepsy is to think about the effect that the sei-
zures have on the:
■■ Quality of life; debate in the literature about whether ‘seizures beget
■■ Brain. seizures’. From human clinical evidence, it is thought
If the seizures are occasional and there is no evi- that treating the first seizures that occur at the onset
dence of increasing frequency or severity, treatment of epilepsy does not actually change the progression
doi:10.1136/inp.d1193 may not be necessary (see Table 1). There is much of the condition. It is therefore unlikely that leaving
seizures untreated affects the progression of epilepsy,
Table 1: Management of recurrent seizures unless a patient has clusters of seizures or has had a
Reasons to treat Reasons not to treat period of status epilepticus (see below).
Animals have one seizure or more per Seizures do not appear to be affecting an
A cluster of seizures is defined as more than one
month animal’s quality of life seizure event in 24 hours. Animals that have clusters
Quality of life is decreased by the seizures, Seizures are infrequent (eg, less than one per of seizures are at risk of going into status epilepticus,
or by the prodromal or postictal periods month) so antiepileptic drug treatment should be instituted
There is a history of clusters of seizures Seizures are very brief and ‘mild’ (eg, focal
or status epilepticus seizures) immediately in such patients.
Seizures increase in frequency or severity The side effects of the antiepileptic drugs are Status epilepticus is defined as prolonged seizure
likely to affect quality of life more than the activity lasting longer than five minutes. It affects neu-
seizures themselves
rons and circuitry within the brain, which predisposes

98 In Practice March 2011 | Volume 33 | 98–104

 Companion animal practice

it to further seizure activity. Hence, any epileptic dog Table 2: Treatment options for epilepsy in dogs
that has a period of status epilepticus should be treated
Drug First-line or add-on drug? Dose
with antiepileptic drugs.
Phenobarbital First-line or add-on 3 mg/kg twice daily
Owner communication is a key part of managing
epileptic dogs. It is vital that owners understand that Potassium bromide First-line or add-on 15 mg/kg twice daily as an add-on drug
or 20 mg/kg twice daily as a first-line drug
antiepileptic drugs control seizures but do not cure
epilepsy. It is also important to be aware of owners’ Levetiracetam Add-on 10 to 20 mg/kg three or four times daily
wishes, and to consider how any treatment and moni- Gabapentin Add-on 10 mg/kg three times daily
toring regimen will fit into their lifestyle. Many own- Zonisamide Add-on 10 mg/kg twice daily
ers have difficulty in understanding that therapy may
not lead to a seizure-free patient, and may still request
treatment even if the seizures are very infrequent. If, and any side effects are more likely to be obvious dur-
for example, seizures occur every three months, giv- ing this period. However, after the first 14 days, side
ing twice daily treatment that may cause side effects is effects often begin to improve.
usually not appropriate. When using phenobarbital, owners should be
Occasionally, owners will decline treatment against informed about:
the veterinary surgeon’s advice. This often occurs if ■■ Its potential side effects, which include:
clients have researched antiepileptic drugs on the ●● Sedation, polyuria, polydipsia and polyphagia;

internet and have learned that drugs such as pheno- ●● Ataxia and paresis;

barbital have a significant risk of causing side effects ●● Hyperexcitability and aggression;

and toxicity. In addition, some owners will be unable ●● Neutropenia, lymphopenia and anaemia (bone

to give twice daily treatment because of their lifestyle. marrow necrosis is uncommon);
Providing there are no signs to suggest that an animal ●● Necrolytic dermatitis (this is uncommon and

is at risk of going into status epilepticus, leaving an epi- does not usually resolve if treatment is ceased);
leptic dog untreated may be acceptable. In such situa- ●● Hepatotoxicity (see Box 2);

tions, it is important to be confident that the animal’s ■■ Metabolic tolerance (hepatic microsomal enzyme
quality of life is not affected by the seizures, and par- induction). When this occurs, the body increases
ticularly that the prodromal and post­ictal periods are its ability to metabolise the drug over time, which
brief and associated with no or only mild behavioural can lead to gradual or sudden increase in seizure
change. Owners need to monitor seizure duration and frequency in dogs that are on an unchanged dose
severity, and the interictal period to ensure that there of phenobarbital, and the dose may need to be
is no evidence of worsening. It would also be prudent increased to keep the same level of seizure control;
for owners to keep diazepam at home, which can be ■■ The need for life-long treatment;
administered per rectum if status epilepticus occurs. ■■ The need for 12-hour dosing;
■■ The importance of not stopping therapy. A sud-
den withdrawal of phenobarbital may cause status
What are the treatment options and epilepticus.
what monitoring is required? Serum concentrations should be checked seven to
14 days after starting therapy or changing the dose. In
Phenobarbital and potassium bromide are both appro-
priate choices for first-line antiepileptic drugs in the
dog (see Table 2). Each drug has advantages and disad-
vantages, and the choice is very much tailored to the
Box 1: Common mistakes when using phenobarbital
individual patient. Phenobarbital needs to be adminis- ■■ Not giving a high enough dose. A starting dose of 3 mg/kg twice daily is sufficient
tered twice daily, while potassium bromide is effective and appropriate for most dogs
when given once daily (although it is tolerated better ■■ Not increasing the dose when the serum concentration is at the low end of the
when the dose is split into two daily doses). Potassium therapeutic range (eg, assuming that 15 mg/ml will be effective when, in fact,
many dogs need a higher serum concentration)
bromide can be difficult to manage if dose changes are
■■ Poor owner compliance/incorrect dosing interval
made frequently due to its long half-life. Each time ■■ Not monitoring serum concentrations
the dose is changed, it takes at least three months to ■■ Abrupt withdrawal of the drug, leading to status epilepticus or more severe seizures
reach a steady state. Potassium bromide is more likely ■■ Giving extra daily doses if a seizure is expected (there is no evidence that this is useful)
to cause vomiting but phenobarbital is hepatotoxic in
some patients, particularly following the long-term
use of high doses. Box 2: Liver function tests
Phenobarbital Dogs on long-term phenobarbital treatment, particularly if serum concentrations are
One of the most common mistakes when using phe- consistently above 35 mg/ml, are at risk of hepatic failure. However, elevation of liver
enzymes, particularly alkaline phosphatase, is quite usual in dogs on phenobarbital and
nobarbital is to start on an inappropriately low dose
is not necessarily a sign of liver dysfunction. It is therefore not useful to use liver enzymes
and to stop treatment abruptly when it has not been in isolation as a test for hepatotoxicity. Instead, measuring pre- and postprandial bile acids
effective (see Box 1). A dose of 3 mg/kg twice daily every six to 12 months, or more frequently if liver dysfunction is specifically suspected in
is sufficient to achieve a therapeutic serum concentra- a patient, is appropriate.
tion in the majority of dogs. It takes seven to 14 days It should also be noted that phenobarbital decreases thyroxine and free thyroxine,
to reach a steady state because the elimination half- and increases thyroid-stimulating hormone (Gaskill and others 2000). However, this is
not associated with clinical signs of hypothyroidism and thyroid supplementation is not
life is approximately 40 to 90 hours. This means that
indicated.
the drug may not start to be effective until that point,

In Practice March 2011 | Volume 33 | 98–104 101

Companion animal practice

addition, they should be measured every three to six phenobarbital. Alternatively, it may be given at a rate
months as a matter of routine, if seizures are poorly of 20 mg/kg twice daily when used as a first-line drug.
controlled or if side effects increase. The therapeutic It is generally very safe to use. It should be given with
range is 15 to 45 mg/ml. However, many epileptic dogs food to avoid vomiting.
will not respond until the serum concentration reaches Side effects include:
at least 25 mg/ml. It is thought that there is a greater ■■ Vomiting;
risk of hepatotoxicity at serum concentrations of more ■■ Ataxia and paresis (which can both be severe);
than 35 mg/ml. Therefore, if 25 to 35 mg/ml is insuf- ■■ Polyuria, polyphagia and polydipsia;
ficient to cause acceptable seizure control, a second ■■ Acute pancreatitis;
drug, usually potassium bromide, should be added. ■■ Megaoesophagus (anecdotal);
The dose required for a desired serum concentra- ■■ Worsening of pruritus in atopic patients.
tion is calculated using the following formula: Although it takes three to six months to reach a
steady stage, measuring the serum concentration at one
Desired serum month and three months provides information about
concentration
New oral dose = Old oral dose x whether the animal is on approximately the right dose.
Current serum
The therapeutic range is 1000 to 3000 mg/ml but, in
concentration
practical terms, the dose can be increased until side
There has been much discussion about whether effects are intolerable or there is clearly no improve-
trough samples (ie, the sample taken when the serum ment in efficacy. Timing of sampling is not important.
concentration is at its lowest) are valuable when moni-
toring phenobarbital concentrations. In most patients,
the timing of sampling does not appear to be impor- When should a second or third
tant, as the serum concentration during steady state antiepileptic drug be added?
does not vary much during the day. However, if a
patient tends to have seizures at a particular time of In patients refractory to phenobarbital and/or potassi-
day, this may suggest breakthrough seizures when the um bromide, polytherapy should generally be avoided,
serum concentration drops at a particular time each if possible. The serum concentration of phenobarbital
day. In such cases, a sample taken at the time of typical and/or potassium bromide should be at the high ends
seizure activity could be very useful. The trough level of the recommended ranges before other drugs are
occurs approximately two hours before the evening added to make sure animals are really resistant to phe-
dose is due. nobarbital and/or potassium bromide. A third drug is
often not particularly effective and owner compliance
Potassium bromide can be a major issue. Owners need to fully understand
Potassium bromide is appropriate as a first-line anti­ that a third drug may not have much effect. However,
epileptic drug in dogs, or it can be used as a second new human drugs are continually being developed and
drug in dogs that have epilepsy that is refractory to some are effective in epileptic dogs.
phenobarbital. It should not be added until the serum
concentration of phenobarbital is 25 to 35 mg/ml. The
main problem associated with the use of potassium Which add-on drug should be used?
bromide is its very long elimination half-life (around
24 days). Therefore, when changing the dose in order Levetiracetam
to alter the serum concentration, it is important to Levetiracetam, given at the recommended dose of 10 to
remember that it can take around three to six months 20 mg/kg three times daily, is arguably the most effec-
to reach a steady state. tive third-line antiepileptic drug in dogs (Volk and oth-
Potassium bromide is a suitable choice for animals ers 2008). It has minimal side effects, which may include
with liver disease as it is renally excreted. It should be sedation, but these are generally rare and mild. It needs
noted that a sudden change in salt content in the diet to be given three times daily or, possibly, four times daily
may affect potassium bromide excretion. For example, in some patients. It has a short half-life (approximately
if a high salt diet is started (eg, some prescription urinary 3·6 hours) and therefore does not reach a steady state.
diets), potassium bromide elimination will increase, However, it can significantly affect seizure frequency
serum concentrations may drop and breakthrough in cases of pharmaco­resistant epilepsy. Levetiracetam is
seizures may occur. Drinking excessive sea water may renally excreted. The main drawback of using levetira-
have a similar effect. Owners should be informed that cetam is that it is usually only efficacious for around six
changing the salt content of an animal’s diet may affect months. In addition, it is expensive.
potassium bromide levels and seizure control.
It should be noted that although potassium bromide Gabapentin
is well tolerated and safe in most dogs, acute pancreati- Gabapentin has also been recommended as an add-on
tis, which is potentially fatal, is a recognised side effect drug in dogs (Platt and others 2006). It is predomi-
in patients given a combination of phenobarbital and nantly renally excreted, but is also partially hepati-
potassium bromide (Gaskill and Cribb 2000). In addi- cally metabolised. It can cause sedation and ataxia,
tion, there is some anecdotal evidence that potassium but this is usually very mild. The recommended start-
bromide is occasionally associated with the develop- ing dose is 10 mg/kg three times daily. It has a very
ment of megaoesophagus. short half-life. Pregabalin, a closely related drug, has
Potassium bromide may be administered at a dose also showed promise in cases of epilepsy (Dewey and
of 15 mg/kg twice daily when used as an add-on with others 2009).

102 In Practice March 2011 | Volume 33 | 98–104

 Companion animal practice

Zonisamide Diagnostic and treatment errors may lead to ‘pseu-

Zonisamide may also be useful as a third drug (von dopharmacoresistance’, and there are several ques-
Klopmann and others 2007). The half-life is typically tions that need to be asked if a patient is apparently
about 15 hours, but it is hepatically metabolised, so pharmacoresistant:
care needs to be taken when using it with phenobarbi- ■■ Has a diagnosis been made?
tal. The recommended dose is 10 mg/kg twice daily. It ■■ Is the diagnosis correct, or is further investigation
may cause sedation and ataxia. required?
■■ Is the treatment correct? That is, is an appropriate
antiepileptic drug being used?
How should the issues of owner ■■ Is the treatment being given appropriately? Here,
compliance, counselling and consideration should be given to owner compliance
communication be handled? issues, dosing regimens (every 12 hours not twice
daily) and dietary changes that could affect drug
Owners can find it difficult to come to terms with being elimination.
told that their dog is epileptic. They often struggle with ■■ Have serum concentrations been measured?
the concept that epilepsy can only be controlled and ■■ If appropriate, have trough levels been measured?
never cured. In addition, generalised seizures are vio- ■■ Has the correct serum concentration been targeted?
lent and can be very distressing to see. Owners often With phenobarbital, a serum concentration of 25
fear that their dog may suffer injury or brain damage to 35 mg/ml (range 15 to 45 mg/ml) may be necessary
during the seizure. Therefore, they typically need regu- for an effective response.
lar veterinary advice and general support. The follow-
ing checklist is useful when discussing epilepsy with
owners to inform them of what to expect and what they How should status epilepticus
need to do to aid a more satisfactory outcome: be treated?
■■ Single seizures lasting one to two minutes are
unlikely to cause significant brain damage; The traditional definition of status epilepticus was
■■ Animals are unlikely to be aware of a seizure; continuous seizure activity lasting more than 30 min-
■■ Stopping antiepileptic drugs suddenly is danger- utes. However, more recently, the more clinically use-
ous, so owner compliance is imperative; ful definition is continuous seizure activity lasting
■■ Clusters of seizures or status epilepticus are life- longer than five minutes.
threatening and veterinary assistance must be In patients experiencing status epilepticus, the most
sought immediately if they occur. urgent priorities are to ensure a clear airway, breathing
Compliance among owners is generally better and circulation, and to stop the seizures. Affected ani-
if they are warned in detail about the side effects of mals should be given intravenous or rectal diazepam
anti­epileptic drugs. These should therefore be dis- at 0·5 to 1 mg/kg, and an intravenous catheter should
cussed with owners or a list provided for them to take be placed immediately. Once the patient has stopped
home. Owners should be encouraged to keep a sei- seizuring, history, physical and neurological examina-
zure diary, which will help them to monitor the effect tion needs to be performed. Other priorities include
of a drug objectively. It will also help the veterinary the measurement of serum glucose and electrolytes.
surgeon to make decisions about drug changes. Rectal temperature should be maintained at between
37 to 39·5°C. Finally, fluid therapy should aim to

How should pharmacoresistant

patients be managed? Box 3: Loading protocols
Loading protocols should be used if serum concentrations need to be raised rapidly.
If phenobarbital and potassium bromide administra- This is recommended in animals that:
tion at doses that achieve therapeutic concentrations ■■ Have presented following a cluster of seizures;
and a steady state does not result in a decrease in seizure ■■ Are not on antiepileptic treatment;
frequency of 50 per cent or more, the patient is deemed ■■ Have recently been treated for status epilepticus;
to have pharmacoresistant epilepsy. In humans, fail- ■■ Have a very high seizure frequency.
Phenobarbital loading is useful for treating status epilepticus. While many loading
ure of seizure control within nine months of treatment
protocols have been published, the author prefers those outlined below.
by a neurologist is one of several definitions that are
used. For these human patients, if all medical manage- Phenobarbital
ment fails, surgery is possible in some cases. This is ■■ Step 1. Administer 3 mg/kg intravenously every 30 minutes until a total dose of
currently not an option for most epileptic dogs and up to 20 mg/kg has been given
cats. ■■ Step 2. Reduce the dose to 3 mg/kg intravenously every six hours for 24 hours
It is estimated that around a third of dogs with ■■ Step 3. Reduce the dose to 3 mg/kg orally twice daily for continued long-term use
Note animals will be very sedated during this time and will need to be hospitalised.
epilepsy do not respond well to phenobarbital and/or
potassium bromide. This may be due to: Potassium bromide
■■ An altered expression of drug transporters in the ■■ Animals can be loaded over five days, with the drug being given orally at a dose of
brain, which pump antiepileptic drugs away from 125 mg/kg/day orally (divided am and pm), and then 20 mg/kg twice daily
where they are needed; or or
■■ Changing drug targets in patients with recurrent ■■ Animals can be loaded over one day, but this requires hospitalisation. In this case, the
dose is given orally at a dose of 100 mg/kg every six hours for 24 hours. A dose should
seizures, which make them resistant to antiepilep-
be skipped if the patient becomes obtunded
tic drugs.

In Practice March 2011 | Volume 33 | 98–104 103

Companion animal praCtiCe

achieve normohydration and normotension to ensure Veterinary surgeons may be under pressure to treat
that cerebral blood flow is maintained. dogs that have very infrequent seizures from owners
Diazepam at a dose of 0·5 to 1 mg/kg should be who are struggling to cope with the risk of any seizures
given up to three times intravenously or rectally. If this occurring at all. Communication and education are
fails to stop the seizures, it is likely that the patient is key, and, with better understanding, many owners can
refractory to diazepam. At this stage, phenobarbital learn to live satisfactorily with an epileptic dog.
should be given as a loading protocol (see Box 3). This
will cause stupor for the first few hours and the patient References and further reading
will need to be hospitalised for several days. Fluids and CHANDLER, K. (2006) Canine epilepsy: what can we learn
oxygen supplementation will be necessary during this from human seizure disorders? Veterinary Journal 172, 207-217
period. DEWEY, C. W., CERDA-GONZALEZ, S., LEVINE, J. M.,
If phenobarbital loading is ineffective, a continuous BADGLEY, B. L., DUCOTÉ, J. M., SILVER, G. M., COOPER,
J. J., PACKER, R. A. & LAVELY, J. A. (2009) Pregabalin as an
rate infusion of propofol can be considered. Propofol
adjunct to phenobarbital, potassium bromide, or a combination
has potent antiseizure properties and an intravenous
of phenobarbital and potassium bromide for treatment of dogs
bolus of 4 to 8 mg/kg can be given to effect. A continu- with suspected idiopathic epilepsy. Journal of the American
ous rate infusion can be given at 6 to 12 mg/kg/hour. Veterinary Medical Association 235, 1442-1449
This regimen should be used with caution. Animals are GASKILL, C. L., BURTON, S. A., GELENS, H. C., IHLE,
likely to be anaesthetised at these doses, so supportive S. L., MILLER, J. B., SHAW, D. H., BRIMACOMBE, M. B.
care including oxygen and fluids will be necessary. If & CRIBB, A. E. (2000) Changes in serum thyroxine and
thyroid-stimulating hormone concentrations in epileptic dogs
the patient is refractory to diazepam and phenobarbi-
receiving phenobarbital for one year. Journal of Veterinary
tal, the prognosis for recovery is guarded. Pharmacological Therapeutics 23, 243-249
GASKILL, C. L. & CRIBB, A. E. (2000) Pancreatitis associated
with potassium bromide/phenobarbital combination therapy in
What seizure frequency is acceptable? epileptic dogs. Canadian Veterinary Journal 41, 555-558
PLATT, S. R., ADAMS, V., GAROSI, L. S., ABRAMSON,
There are no definite answers to this question. The C. J., PENDERIS, J., DE STEFANI, A. & MATIASEK, L.
(2006) Treatment with gabapentin of 11 dogs with refractory
effect of epilepsy on quality of life will depend on the
idiopathic epilepsy. Veterinary Record 159, 881-884
severity and length of the seizures, and the character THOMAS, W. B. (2010) Idiopathic epilepsy in dogs and cats.
of the prodromal and postictal phases. Some dogs are Veterinary Clinics of North America: Small Animal Practice
very anxious or aggressive either pre- or post-seizure, 40, 161-179
and some animals may be temporarily blind or pro- VOLK, H. A., MATIASEK, L. A., LUJÁN FELIU-PASCUAL,
foundly ataxic following a seizure. However, in certain A., PLATT, S. R. & CHANDLER, K. E. (2008) The efficacy and
tolerability of levetiracetam in pharmacoresistant epileptic dogs.
patients, the quality of life may still be reasonable even
Veterinary Journal 176, 310-319
with one seizure occurring every two weeks or more.
VON KLOPMANN, T., RAMBECK, B. & TIPOLD, A.
Euthanasia is often considered if animals are having (2007) Prospective study of zonisamide therapy for refractory
weekly seizures in the face of aggressive treatment. idiopathic epilepsy in dogs. Journal of Small Animal Practice
In such cases, owner communication is imperative. 48, 134-138

104 In Practice March 2011 | Volume 33 | 98–104

Copyright of In Practice (0263841X) is the property of BMJ Publishing Group and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.