Seizure disorders are common in dogs, the most common etiology being idiopathic epilepsy. Managing seizure disorders presents a major challenge to the veterinarian, especially when a dog does not respond to standard (i.e., phenobarbital, bromide) therapy. Such refractory cases account for between 25–30% of all epileptics. It is very important for the clinician to inform the pet owner that most epileptic dogs do not reach seizure-free status; success is typically considered a reduction in the frequency and duration of seizures. Nonetheless, the goal of anticonvulsant therapy should be to eliminate seizure activity in the patient, or come as close to this goal as possible, without subjecting the patient to unacceptable side effects of drug therapy or the client to unreasonable financial burden.

Zonisamide has been shown to be effective for the treatment of both focal and generalized seizures in people, with minimal side effects. Zonisamide (ZNS) is metabolized mainly by hepatic microsomal enzymes. The t ½ in dogs is roughly 15–20 hrs and in cats is about 33 hrs. In humans, it has been shown that the t ½ of ZNS is dramatically shorter in patients already receiving drugs that stimulate hepatic microsomal enzymes, in comparison with patients who are not receiving such drugs. A similar phenomenon appears to occur in dogs. When used as an add-on therapy for dogs already receiving drugs requiring hepatic metabolism (e.g., phenobarbital), I recommend an initial oral ZNS dose schedule of 8–10 mg/kg body weight, q 12 hrs. This dose regimen has been shown to maintain canine serum ZNS concentrations within the therapeutic range reported for people (10 to 40 ug/ml), when used as an add-on therapy. For dogs not concurrently receiving drugs that induce hepatic microsomal enzymes, it is recommended to start ZNS at a dosage of 5 mg/kg body weight, q 12 hrs. Trough serum ZNS concentrations are checked after approximately one week of treatment. Zonisamide has a high margin of safety in dogs. In one clinical trial, ZNS was found to decrease seizure frequency by at least 50% in 7 of 12 dogs with refractory idiopathic epilepsy. In this responder group, the mean reduction in seizure frequency was 81.3%. In six of the 7 responder dogs, phenobarbital was reduced by an average of 92.2%. Mild side effects (e.g., transient sedation, ataxia, vomiting) occurred in six (50%) dogs; none of the side effects were considered severe enough to discontinue zonisamide therapy. In another similar study, 9 of 11 refractory epileptic dogs treated with zonisamide were responders, with a median seizure reduction of 92.9%; transient ataxia and sedation occurred in six dogs. I have used zonisamide as a sole anticonvulsant drug in a large number of dogs. Zonisamide is usually effective as a sole anticonvulsant therapy, with few to no apparent side effects. I have also treated a number of cats with zonisamide for seizure control. The elimination half-life of ZNS in cats is quite long, so some cats can be dosed SID. The case numbers are small, but the drug does appear to be of some clinical utility in this species. I have had a few cats become anorexic on ZNS, necessitating drug discontinuation.

Felbamate has shown efficacy for both focal (partial) and generalized seizures in experimental animal studies and human clinical trials. There is also evidence that felbamate may afford some protection to neurons against hypoxic/ischemic damage. Approximately 70% of the orally administered dose of felbamate in dogs is eliminated in the urine unchanged; the remainder undergoes hepatic metabolism. The t ½ of felbamate in adult dogs is typically between 5 and 6 hrs (range, 4–8 hrs). An initial felbamate dose regimen of 15–20 mg/kg body weight, q 8 hrs is recommended. Felbamate has a wide margin of safety in dogs. If the initial dose of felbamate is ineffective, I increase the dose in 15 mg/kg every 2 weeks until efficacy is achieved, unacceptable side effects are evident, or the drug becomes cost-prohibitive. A major advantage of felbamate over more standard anticonvulsant drugs is that it does not cause sedation. Because felbamate does undergo some hepatic metabolism, liver dysfunction is a potential side effect. Reversible blood dyscrasias and KCS have also been associated with felbamate use. In recent years, felbamate has increased substantially in price, so I have been using it very infrequently.

Levetiracetam (LEV) has demonstrated efficacy in the treatment of focal and generalized seizure disorders in people, as well as in several experimental animal models. The mechanism of action for levetiracetam's anticonvulsant effects is not entirely clear, but appears to be related to its binding with a specific synaptic vesicle protein (SV2A) in the brain; unlike other anticonvulsant drugs, levetiracetam does not appear to directly affect common neurotransmitter pathways (e.g., GABA, NMDA) or ion channels (e.g., sodium, T-type calcium). Levetiracetam has demonstrated neuroprotective properties, and may ameliorate seizure-induced brain damage. Levetiracetam has also been reported to have an "anti-kindling" effect, which may diminish the likelihood of increasing seizure frequency over time. Orally administered LEV is approximately 100% bioavailable in dogs, with a serum t ½ of 3–4 hours. Approximately 70% of the administered dose of LEV is excreted unchanged in the urine; the remainder of the drug is hydrolyzed in the serum and other organs. There does not appear to be any hepatic metabolism of LEV in either humans or dogs. The effective serum LEV concentration in people is 5 to 45 ug/ml. Since there is no clear relationship between serum drug concentration and efficacy for LEV, and since the drug has an extremely high margin of safety, routine therapeutic drug monitoring is not typically recommended for this drug in people. An initial dosing schedule of 20 mg/kg body weight, q 8 hours has been recommended for dogs and cats. This dose can be increased by 20 mg/kg increments until efficacy is achieved, side effects become apparent, or the drug becomes cost-prohibitive. Long-term toxicity data for LEV in dogs confirm that the drug is extremely safe. Despite initial reports of efficacy for dogs as an add-on drug, recent evidence has suggested that LEV is often an ineffective maintenance add-on drug in dogs and also may have a substantial "honeymoon effect". However, the failure of LEV in dogs may be somewhat due to inadequate plasma drug concentrations; LEV elimination has recently been shown to be accelerated by concurrent PB administration in dogs. My colleagues and I prospectively investigated the use of oral LEV as an add-on anticonvulsant therapy for epileptic cats refractory to phenobarbital. LEV appears to be very well tolerated in this species, usually with no apparent side effects, and no obvious "honeymoon effect". The t ½ of elimination is approximately 3 hours after oral administration. A dose of 20 mg/kg PO, q 8 hrs typically achieves a serum drug level within the therapeutic range reported for people. Two of 12 cats experienced transient inappetance and lethargy that resolved without dose adjustment within 2 weeks. Although there is some degree of variability among cats, the mean reduction of seizure frequency in cats receiving levetiracetam as an add-on drug is approximately 68%; this was found to be statistically significant when compared to the pre-levetiracetam time period. In addition, 7 of 10 cats evaluated for seizure frequency reduction were responders (i.e., reduction of seizure frequency of 50% or more), with a mean reduction of seizures of 92%. I consider LEV to be the preferred add-on anticonvulsant drug for cats receiving PB, due to lack of serious side effects and evidence of efficacy.

Gabapentin is a structural analog of GABA. Gabapentin is thought to exert its anti-seizure effects via binding to the α2δ subunit of voltage-gated neuronal calcium channels. This binding decreases intracellular calcium influx, leading to decreased synaptic release of excitatory neurotransmitters. Gabapentin is well absorbed in both dogs and people, with peak serum concentrations occurring within 1–3 hrs after ingestion. In people, virtually all the orally administered dose of gabapentin is excreted unchanged in the urine (i.e., no hepatic metabolism). In dogs, however, 30–40% of the orally administered dose of gabapentin undergoes hepatic metabolism to N-methyl-gabapentin. Despite undergoing some hepatic metabolism in dogs, there is no appreciable induction of hepatic microsomal enzymes in this species. The t ½ for gabapentin in dogs and cats is between 3 and 4 hrs. The recommended dose range of gabapentin for dogs is 25–60 mg/kg body weight, divided, q 6–8 hrs. I use an initial dose regimen of 10 mg/kg body weight, q 8 hrs. The suspected therapeutic range for dogs is 4–16 mg/L. As is the case with felbamate, serum gabapentin concentrations are seldom pursued in dogs. Sedation does not appear to be a common problem with gabapentin use in dogs, but occasionally occurs. There are two clinical reports of gabapentin use as an add-on drug for dogs with refractory epilepsy. Overall, the responder rate of these dogs was between 41% and 55%. Because of its short t ½, gabapentin probably needs to be administered at least every 8 hrs, and possibly every 6 hrs, in order to maintain serum gabapentin concentrations within the therapeutic range. The potential need for q 6 hr dosing can make it difficult for some pet owners to reliably administer gabapentin.

A newer gabapentin analog, pregabalin, has recently been approved for human use. Pregabalin has an increased affinity for the α2δ-subunit of voltage-gated calcium channels, compared with gabapentin, and is suspected to be more effective in people than its predecessor as both an anticonvulsant and a pain-relieving drug. My colleagues and I have completed an oral pharmacokinetic study of pregabalin in both dogs and cats, and a clinical trial of pregabalin in refractory epileptic dogs. The median seizure reduction in the clinical study was approximately 50%. The half-life of elimination of pregabalin in dogs is about 7 hours and in cats it is about 11 hrs. We have been giving pregabalin to dogs at a starting dose of 2 mg/kg q 12 hrs, and increasing it if needed by 1 mg/kg increments up to 4 mg/kg q 12 hrs. Our experience with pregabalin in cats is still limited, but the starting dose we are using for this species is 1–2 mg/kg q 12 hrs.