The International Veterinary Epilepsy Task Force has published guidelines for the initiation of anticonvulsant treatment in dogs² that can be extrapolated for cats. The task force recommends initiation of anticonvulsant therapy if a dog (or cat) has ≥2 seizures in 6 months and/or a single seizure that lasts >5 minutes.^2,3 These guidelines are written to aid clinicians in decision-making for epileptic patients; however, individual differences (eg, metabolic state, severe adverse effects experienced during or after the seizure) may drive earlier initiation of anticonvulsant therapy than outlined.

The presence of postictal signs that are especially severe (eg, aggression) or long-lasting (ie, >24 hours) is a secondary reason to consider anticonvulsant therapy. The author also encourages initiation of anticonvulsant therapy if the patient is at high risk for additional seizures but the frequency recommendations are not met. For example, anticonvulsant therapy may be considered for a young dog with clinical hydrocephalus in which one seizure has already been observed.

Despite decades of research on genetic factors, identification of specific seizure syndromes, and development of novel anticonvulsants, selection of appropriate anticonvulsants by clinicians is still largely based on seizure semiology (ie, the appearance, or phenotype, of the seizure), patient health status, pet owner considerations (eg, frequency of administration, cost, formulation type), and the clinician’s comfort with anticonvulsants.

Monotherapy is typically recommended over polytherapy during the initial treatment period. Starting one anticonvulsant drug and assessing the clinical response—including clinical adverse effects, effect on seizure control, and, if applicable, serum drug concentrations—is recommended long enough for attainment of steady state and an observation period. The author recommends observing for a minimum of 3 seizure cycles before changing the drug protocol. Exceptions to this include patients with severe adverse clinical effects. Initiation with polytherapy is more often employed when using potassium bromide without a loading period. In this situation, potassium bromide takes up to 12 weeks to reach steady state; therefore, a “bridge” drug is used to improve seizure control until bromide is at steady state. After attaining steady state, the bridge drug can be tapered or removed if indicated.

Due to their relatively high frequency of use and availability in the United States, this article focuses on phenobarbital, potassium bromide, levetiracetam, and zonisamide.

Phenobarbital has been widely used for seizure management since the 1900s and is effective for seizure control in both dogs and cats.^2,4 The anticonvulsant effect of phenobarbital is likely due to increased intracellular chloride secondary to prolonged opening of the γ-aminobutyric acid receptors on the postsynaptic membrane. Increased intracellular chloride causes increased negative membrane potential, which reduces nerve firing.⁵

Adverse effects are rare in cats but common in dogs⁶ and can include sedation, ataxia, weakness, polydipsia, polyphagia, and polyuria. Hepatopathy and hepatotoxicity have been well-documented in dogs receiving phenobarbital. Hepatopathy/toxicity has not been reported in cats receiving phenobarbital long-term; therefore, elevated ALP or ALT in cats receiving phenobarbital should prompt investigation into other hepatopathies.⁴ Clinical adverse effects and risk for hepatopathy are serum phenobarbital-dependent; therefore, maintaining serum phenobarbital concentrations <35 μg/mL can be targeted to reduce these adverse effects. Conversely, blood dyscrasias are idiosyncratic and not dependent on serum concentration. Blood dyscrasias have been reported in 4.2% of dogs receiving phenobarbital. If anemia, thrombocytopenia, leukopenia, or pancytopenia are noted, phenobarbital should be safely and swiftly removed from administration and replaced by another anticonvulsant.⁷ CBC monitoring is recommended 14 days after starting phenobarbital and every 6 months thereafter.

A typical starting dosage for dogs and cats is 2-5 mg/kg PO every 12 hours (reported range, 1.8-10 mg/kg/day).^2,4,6,8-10 Steady state is expected 10 to 14 days after initiation of treatment in both species.^11,12 The reference interval for monitoring serum concentrations at steady state levels for dogs is 15 to 40 µg/mL; however, the author agrees that optimal therapeutic success is frequently obtained with serum phenobarbital concentrations of 25 to 30 µg/mL, which can help limit toxicity.^4,9-11 

A therapeutic reference interval for cats is not available, but seizure control was achieved in a study in 93% of cats with serum phenobarbital concentrations of 15 to 45 µg/mL, regardless of the underlying cause.⁴ Transdermal phenobarbital has been investigated for use in cats.^13,14 In these studies, serum phenobarbital concentrations in healthy cats were within the therapeutic reference range for dogs^13,14; however, a prospective clinical trial in epileptic cats showed the serum concentrations did not correlate with the dose administered.¹⁵ Additional routes of administration in cats are being explored. Clinicians are encouraged to monitor forthcoming literature for additional options beyond per os administration of phenobarbital.

Bromide has well-documented seizure control success in dogs.² In a study, there was a strong level of evidence for use of potassium bromide as a monotherapy and weak evidence for use as adjunct therapy.⁷ Another study reported bromide to be a reasonable first-line choice for dogs with seizures but found that it was slightly less effective than phenobarbital in the first 6 months of therapy.¹⁶ In the author’s clinic, bromide is preferentially used in dogs with cluster seizures and focal seizures.

Bromide is a halide salt that is thought to mimic chloride and thereby hyperpolarize neuronal membranes and result in seizure control. Due to chloride mimicry, serum chemistry analyzers may misidentify bromide as chloride, which can cause falsely elevated chloride on serum chemistry results. Clinicians should take care to differentiate mimicry and actual changes in chloride. Further, dietary chloride may affect serum bromide concentrations. Lowering the chloride content in the body can lead to an increase in bromide and vice versa.^17,18 Dogs do not require a specific diet during bromide treatment but should be provided a chloride-stable diet (ie, a dog receives the same food, treats, and access to human food every day, without change, to avoid fluctuations in chloride).

The starting dosage of bromide is 40 mg/kg PO once or twice (divided dose) daily. Time to steady state is ≈12 weeks, and a therapeutic reference interval for dogs is available. Serum bromide concentrations of 0.88-3 g/L are considered to be within the acceptable reference interval.^11,12,19 Common adverse effects include polyuria, polydipsia, increased appetite, and neurologic signs (eg, ataxia, sedation, weakness).²⁰ Pancreatitis has commonly been associated with bromide administration, but whether this is a primary adverse effect of bromide or secondary to polyphagia remains unclear. 

Bromide is not recommended for use in cats. Reversible neutrophilic and eosinophilic lower airway disease have been documented in cats receiving bromide for seizure management.^21,22

Levetiracetam has a novel mechanism of action as compared with other common anticonvulsants.^23,24 The therapeutic reference interval of levetiracetam for dogs and cats is unknown and has been extrapolated from that of humans (5-45 µg/mL).^23,25,26 The most commonly reported adverse effects include vomiting, sedation, hypersalivation, ataxia, and hyperactivity.^7,23,24 Adverse clinical effects have been reported in >30% of studies; however, reports of toxicity are not well-documented.⁷

Two formulations of levetiracetam are available: intermediate-release and extended-release. The starting dosage for intermediate-release levetiracetam is 20 mg/kg PO every 8 hours (dogs and cats)²³ and for extended-release levetiracetam is 30 mg/kg PO every 12 hours for dogs ≥33 lb (15 kg) and 500 mg PO once daily for cats ≥11 lb (5 kg).^27-29 When administering 500-mg extended-release levetiracetam to a cat, it is important to give the entire tablet once daily and not to crush, split, or allow the cat to chew the tablets, any of which could nullify the release effect. In dogs, administration of extended-release levetiracetam with food typically results in a lower maximal serum concentration as compared with that in fasted dogs.²⁵ In one study, rectal levetiracetam at 40 mg/kg successfully stopped active seizures in most dogs.³⁰ Rectal levetiracetam may be an alternative to rectal diazepam for home use in some epileptic patients.

Loss of efficacy over time has been suggested to occur in dogs but has not been documented in cats.^7,31 However, the lack of documentation does not exclude the possibility of long-term tolerance in cats. Administration of intermittent or pulse levetiracetam for several days has been recommended for cluster seizure management.³² This approach has not been validated in cats and therefore should be employed with caution. 

Levetiracetam has been championed for use in specific seizure syndromes in human and veterinary medicine. Specifically, reflex seizures in dogs and cats and myoclonic seizures in humans demonstrate a marked response to levetiracetam as compared with other anticonvulsants.^33-36 As genetic factors that influence seizures are discovered, syndrome-specific anticonvulsant recommendations will become more common.

Zonisamide is a sulfonamide-derived anticonvulsant developed for use in cats and rats in the late 1970s in Japan.³⁷ According to the 2015 Small Animal Consensus statement, there is a low level of evidence supporting the use of zonisamide in cats and dogs.^2,38 The starting dosage for dogs and cats is 10-20 mg/kg PO every 12 hours; a therapeutic serum reference interval of 10 to 40 µg/mL has been reported in dogs.³⁹ The reported half-life in cats is longer than in dogs, which is suspected to be due to the decreased hepatic glucuronide conjugation in cats.⁴⁰

In one study, 50% of cats receiving zonisamide at 20 mg/kg/day had GI adverse effects; thus, a lower dose may be required.⁴⁰ Several small studies have reported seizure control in 60% to 80% of dogs receiving zonisamide at 7-10 mg/kg PO twice daily, with 40% to 60% of dogs showing 1 or more adverse clinical effect.^7,39,41 The most frequently reported adverse effects in dogs and cats include ataxia, sedation, and GI upset (eg, vomiting, diarrhea, nausea).^7,40 Elevations in ALT and ALP, with rare idiosyncratic hepatotoxicity and renal tubular acidosis, have been well-documented.^42,43 In a study of 107 dogs, acute clinical and serum chemical hepatopathy was reported to be low (<1%). The study authors suggested that monitoring for changes in liver enzymes within 4 weeks of initiating therapy should detect acute hepatopathy.⁴⁴ 

Renal tubular acidosis has been reported with zonisamide use in dogs and humans.⁴³ Monitoring for this rare adverse effect should include blood gas testing and urinalysis. Care should be taken to avoid concurrent administration of bromide and zonisamide. Clinicians should inform owners of the limited published data when prescribing zonisamide. 

Anticonvulsants are the mainstay treatment for dogs and cats with epilepsy. The clinical benefits, adverse effects, and disposition of phenobarbital, bromide, and levetiracetam are well-reported in the veterinary literature. Therefore, these drugs—with the exception of levetiracetam in cats—may be recommended as first-line therapy. Fewer studies have been published regarding the clinical benefits, adverse effects, and disposition of zonisamide. Despite the recent popularity of zonisamide, clinicians should continue to disclose its limitations to owners until more data are available. Regardless of the anticonvulsant chosen, appropriate monitoring and awareness of clinical adverse effects are critical for long-term management of epileptic patients.