Materials and Methods

The study was conducted as an open-label, noncomparative clinical trial. Cats with generalized or partial (focal) seizures that were suspected to have idiopathic epilepsy and were currently receiving phenobarbital were eligible for enrollment in the trial if they were still having seizures despite treatment with phenobarbital, they had developed unacceptable adverse effects associated with phenobarbital treatment, or they had unacceptably high serum phenobarbital concentrations. In addition, cats were eligible for enrollment in the trial only if they had had at least 2 seizures prior to enrollment.

For cats considered for enrollment in the study, a presumptive diagnosis of idiopathic epilepsy was made if seizures had been observed, but results of physical and neurologic examinations performed during the interictal period were normal, the owner reported that the cat did not have any behavioral abnormalities at home, and results of routine laboratory testing and diagnostic imaging of the brain (ie, computed tomography and magnetic resonance imaging), when performed, were normal. Owners of all cats enrolled in the study provided written consent.

Cats eligible for enrollment in the study were examined by 1 or more of the authors, at which time a complete history was obtained and physical and neurologic examinations were performed. In particular, information was obtained pertaining to the onset of seizures, frequency of seizures, general behavior and behavior during the interictal period, anticonvulsant drug treatment (ie, drugs administered and dosages), response to treatment with anticonvulsant drugs (including reason for treatment failure), and adverse effects associated with anticonvulsant drug treatment. Seizure logs maintained by the owners were obtained, when available, and reviewed. Additional information that was obtained included breed, sex, body weight, age at the onset of seizures, current age, results of previous physical and neurologic examinations, results of previous diagnostic testing, and serum anticonvulsant drug concentrations (when available).

Cats were treated with levetiracetam^c at a dosage of 20 mg/kg (9.1 mg/lb), PO, every 8 hours. The calculated dose was rounded to the nearest whole number, and the drug was compounded at a private pharmacy. Owners were instructed to record in a log book the date and a description of any seizure episodes that occurred during treatment with levetiracetam. Cats were followed up for a minimum of 3 months after treatment with levetiracetam was initiated. Phenobarbital administration was continued, and dosage changes were made only as needed on the basis of serum phenobarbital concentration and seizure control. No additional anticonvulsant drugs were added to the treatment regimen during the study period.

After a minimum of 1 week of treatment, serum levetiracetam concentrations were measured immediately before and 2, 4, and 6 hours after administration of a drug dose. Blood samples were collected, and serum was harvested within 2 hours and shipped overnight on ice to a commercial laboratory,^d where serum levetiracetam concentrations were measured by means of highperformance liquid chromatography, as described.^21,22

Data analysis—Log books detailing seizures that occurred before and after initiation of levetiracetam administration were obtained from owners of cats enrolled in the study and reviewed. For each cat, the owner log and medical record were reviewed and the owner and referring veterinarian were contacted to determine the numbers of seizures that occurred before and after initiation of levetiracetam treatment. Seizure frequencies before and after initiation of levetiracetam treatment were then calculated as number of seizures per month. When seizures occurred in clusters, individual seizures were counted separately if they were clearly documented by the owner or in the medical record. Alternatively, each cluster episode was counted as a single seizure if individual seizures were not recorded. Status epilepticus was defined as a single seizure episode lasting > 5 minutes, and each episode of status epilepticus was counted as a single seizure.

The Wilcoxon signed rank test was used to compare seizure frequency prior to levetiracetam treatment with seizure frequency after initiation of levetiracetam treatment. Percentage reduction in seizure frequency was calculated by subtracting seizure frequency after initiation of levetiracetam treatment from seizure frequency prior to levetiracetam treatment and dividing by seizure frequency prior to levetiracetam treatment. Cats were considered to have responded to levetiracetam treatment if the percentage reduction in seizure frequency was ≥ 50%. Finally, the Wilcoxon signed rank test was used to compare total number of seizures that occurred during the 3 months immediately prior to initiation of levetiracetam treatment with total number of seizures that occurred during the 3 months immediately after initiation of levetiracetam treatment; only cats for which information for the 3 months before and after initiation of levetiracetam treatment was available were included. Analyses were performed with standard software.^e Values of P < 0.05 were considered significant.

Safety and tolerability of levetiracetam were determined by monitoring cats for adverse effects during follow-up examinations by 1 or more of the authors and during telephone conversations with owners and referring veterinarians. In addition, a CBC and serum biochemical profile were performed at the same time that serum levetiracetam concentrations were measured.

Serum levetiracetam concentrations were analyzed by means of noncompartmental pharmacokinetic methods; standard software was used.^f Area under the concentration-versus-time curve and area under the first moment curve were determined to the last time point by means of the linear trapezoidal method, and time to maximum serum concentration and maximum serum concentration were obtained from examination of the concentration-versus-time curve. Clearance was calculated as dose divided by area under the concentration-versus-time curve, volume of distribution was calculated as dose divided by maximum serum concentration, and mean residence time was calculated as area under the first moment curve divided by area under the concentration-versus-time curve. Because serum concentrations were not measured after IV administration of levetiracetam, true clearance and volume of distribution could not be calculated. Therefore, clearance and volume of distribution are reported as a function of bioavailability.

Results

Twelve cats met the criteria for enrollment in the study. Eleven of the 12 cats were domestic shorthaired cats, and 1 was a Siamese. Three were spayed females, and 9 were castrated males. Median body weight at the time of enrollment was 5.0 kg (11 lb; range, 2 to 7.3 kg [4.4 to 16.1 lb]). Median age at the onset of seizures was 2 years (range, 0.25 to 16 years), and median age at the time of enrollment was 4.3 years (range, 1 to 17 years).

In 11 of the 12 cats, results of a physical examination performed at the time of study enrollment were normal. The remaining cat had ocular and nasal discharges and inspiratory stridor consistent with upper respiratory tract infection. In 10 of the 12 cats, results of a neurologic examination performed at the time of study enrollment were normal. One cat had decreased menace responses; this cat had had multiple seizures immediately prior to examination, and results of a follow-up neurologic examination were normal. The remaining cat was in status epilepticus at the time of initial examination, but results of a follow-up neurologic examination performed 1 week after discharge from the hospital were normal.

Extent of diagnostic testing prior to study enrollment varied among cats depending on participating institution, owner compliance and financial constraints, neurologic status at the time of initial examination, and history. Five of the 12 cats underwent magnetic resonance imaging of the brain. Results were normal in 4 of the 5 cats. In the remaining cat, a fluidfilled cavity was seen in the left pyriform lobe; this cavity was contiguous with the subarachnoid space. There was no evidence of parenchymal compression, and magnetic resonance characteristics of the fluid were consistent with CSF. The lesion was considered to most likely represent a subarachnoid cyst. Two of the 12 cats underwent computed tomography of the brain, and results were normal in both. In 7 cats, CSF was analyzed. Results were normal in 6 of the 7. The remaining cat had a high WBC count and protein concentration consistent with mild suppurative inflammation. Cerebral spinal fluid from 2 cats was tested for evidence of coronavirus infection, and results were negative in both cats. Results of a CBC and serum biochemical profile performed in all 12 cats were unremarkable. Five cats were tested for FeLV and FIV infection (results were negative for all 5), 4 cats were tested for IgG and IgM against Toxoplasma spp (all 4 were seronegative), and 4 cats were tested for serum bile acids concentrations before and after eating (concentrations were within reference limits).

Median duration of phenobarbital treatment prior to enrollment in the study was 7 months (range, 1 week to 46 months). Median phenobarbital dosage was 3.02 mg/kg (1.37 mg/lb), PO, every 12 hours (range, 0.75 to 4.9 mg/kg [0.34 to 2.23 mg/lb], PO, q 12 h). Serum phenobarbital concentration was measured in 10 cats, and median concentration was 29 μg/mL (range, 5.6 to 75 μg/mL; reference range, 15 to 45 μg/mL).

Ten of the 12 cats continued to have seizures while being treated with phenobarbital, including 1 cat that had an episode of status epilepticus and 2 cats that had episodes of cluster seizures. Serum phenobarbital concentration was not measured in 2 of these 10 cats, but in 4 cats, serum phenobarbital concentration was within reference limits (median, 29 μg/mL; range, 24.9 to 34 μg/mL), and in the other 4, serum phenobarbital concentration was high (median, 45.9 μg/mL; range, 37 to 52.9 μg/mL). Three cats, including 1 of the 2 cats with cluster seizures, reportedly developed adverse effects while being treated with phenobarbital, including ataxia and lethargy. One of these cats also had seizures despite having a serum phenobarbital concentration of 26.9 μg/mL. A second cat had a serum phenobarbital concentration of 39.9 μg/mL and did not have any seizures. The cat with cluster seizures had a serum phenobarbital concentration of 34 μg/mL.

Prior to levetiracetam treatment, 3 of the 12 cats were treated with other anticonvulsant drugs in addition to phenobarbital. This included 1 cat treated with potassium bromide; 1 treated with gabapentin; and 1 treated with a combination of potassium bromide, zonisamide, and phenytoin.

Median starting dosage of levetiracetam was 23.6 mg/kg (10.7 mg/lb), PO, every 8 hours (range, 17.2 to 34.7 mg/kg [7.8 to 15.8 mg/lb], PO, q 8 h). Eleven of the 12 cats were still receiving levetiracetam at the time the clinical trial was discontinued, with follow-up times ranging from 6 to 24 months (median, 9 months).

Accurate seizure logs were not maintained for 2 cats. Therefore, efficacy data were available for only 10 cats. For these 10 cats, median seizure frequency prior to treatment with levetiracetam (2.1 seizures/mo; range, 0.8 to 42.4 seizures/mo) was significantly (P = 0.01) higher than median seizure frequency after initiation of levetiracetam treatment (0.42 seizures/mo; range, 0 to 1.25 seizures/mo; Figure 1). Seven of the 10 cats were considered to have responded to levetiracetam treatment (ie, ≥ 50% reduction in seizure frequency); this included 3 cats in which no seizures were seen after treatment with levetiracetam was begun (100% reduction in seizure frequency). Median percentage reduction in seizure frequency for cats classified as responders was 92% (range, 77% to 100%). Percentage reductions in seizure frequency for the 3 cats that did not respond to levetiracetam treatment were 4%, 12.5%, and 34.9%.

Seizure frequency (No. of seizures/mo) in 10 cats with idiopathic epilepsy before (black bars) and after (white bars) addition of levetiracetam to the treatment regimen. Numbers above the bars indicate months of observation prior to initiation of levetiracetam treatment/months of observation after initiation of levetiracetam treatment.

Citation: Journal of the American Veterinary Medical Association 232, 6; 10.2460/javma.232.6.867

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Seizure frequency (No. of seizures/mo) in 10 cats with idiopathic epilepsy before (black bars) and after (white bars) addition of levetiracetam to the treatment regimen. Numbers above the bars indicate months of observation prior to initiation of levetiracetam treatment/months of observation after initiation of levetiracetam treatment.

Citation: Journal of the American Veterinary Medical Association 232, 6; 10.2460/javma.232.6.867

• Download Figure
• Download figure as PowerPoint slide

Seizure frequency (No. of seizures/mo) in 10 cats with idiopathic epilepsy before (black bars) and after (white bars) addition of levetiracetam to the treatment regimen. Numbers above the bars indicate months of observation prior to initiation of levetiracetam treatment/months of observation after initiation of levetiracetam treatment.

Citation: Journal of the American Veterinary Medical Association 232, 6; 10.2460/javma.232.6.867

• Download Figure
• Download figure as PowerPoint slide

For 7 cats, information was available on the number of seizures that occurred during the 3 months immediately before and the 3 months immediately after initiation of levetiracetam treatment, including 4 cats that responded to levetiracetam treatment and 3 that did not. For these 7 cats, number of seizures during the 3 months immediately before initiation of levetiracetam treatment was not significantly (P = 0.109) different from number of seizures during the 3 months after initiation of levetiracetam treatment.

Two of the 12 cats reportedly developed mild adverse effects in association with levetiracetam treatment. One cat had mild inappetence, and the other had mild lethargy and inappetence. These signs resolved without any change in medications after approximately 2 weeks and 1 week, respectively. Results of CBCs and serum biochemical profiles performed at the same time serum levetiracetam concentrations were measured were within reference limits in 11 of the 12 cats. The remaining cat had mild anemia (PCV, 26.7%; reference range, 28% to 48%); however, PCV was within reference limits on follow-up testing.

Administration of levetiracetam was discontinued in 1 cat in which seizure frequency was reduced by only 4%. This cat also had signs of mild cerebellar ataxia while being treated with phenobarbital and levetiracetam. The treatment regimen was changed to include zonisamide, and seizures were reportedly well controlled, although the ataxia persisted.

In 8 of the 12 cats, serum levetiracetam concentration was highest 2 hours after drug administration, and in 2, serum concentration was highest 4 hours after drug administration. Median maximum serum levetiracetam concentration was 25.5 μg/mL (Table 1). Elimination half-life was calculated as 19 hours in 1 cat but was < 6 hours in the remaining 11 cats (median, 2.9 hours).

Discussion

Results of the present study suggested that levetiracetam may be an effective anticonvulsant when given as an adjunct to phenobarbital treatment in cats with idiopathic epilepsy and that serum drug concentrations similar to those reported to be effective in human patients with epilepsy²³ could be achieved with an 8-hour dosing interval. Seven of 10 cats for which efficacy data were available had a ≥ 50% reduction in seizure frequency after treatment with levetiracetam was initiated, including 3 cats that reportedly did not have any seizures after treatment with levetiracetam was begun, and seizure frequency after treatment with levetiracetam was initiated was significantly lower than frequency prior to levetiracetam treatment.

For the 7 cats in the present study for which information was available, number of seizures during the 3 months immediately before initiation of levetiracetam treatment was not significantly different from number of seizures during the 3 months after initiation of levetiracetam treatment. However, we do not believe that this was a useful measure of the effectiveness of levetiracetam in these cats. Cats were included in the present study if they had an increase in seizure frequency while being treated with phenobarbital, developed adverse effects secondary to phenobarbital administration, or had high serum phenobarbital concentrations with or without adequate seizure control. In particular, cats were not required to have had seizures during the preceding 3 months to be eligible for inclusion, and 2 of the 3 cats that were excluded from this analysis had had severe seizures that appeared to be refractory to phenobarbital treatment alone but did not have any additional seizures after treatment with levetiracetam was begun. One cat that was included in this analysis was given levetiracetam not because of poor seizure control, but because of a high serum phenobarbital concentration (52.9 μg/mL). Unfortunately, the cat's owner discontinued phenobarbital treatment after treatment with levetiracetam was begun, and the cat had an increase in seizure frequency. Treatment with phenobarbital was reinstituted at a lower dosage, and within about 6 months, the cat's seizure frequency again decreased.

Long-term follow-up is needed to evaluate the efficacy of levetiracetam because in dogs and people, the drug has been associated with a transient decrease in seizure frequency with a subsequent return to baseline values (“honeymoon” period).^19,g In dogs, this transient period of decrease in seizure frequency may last 4 to 8 months.^g Of the 7 cats in the present study that were followed up for ≥ 8 months, 5 were classified as responding to treatment. However, additional study is needed to determine whether levetiracetam will result in only transient decreases in seizure frequency in cats.

Calculated reductions in seizure frequency in the present study may have been conservative because of the manner in which seizure episodes were counted. So as to not overestimate seizure frequency prior to levetiracetam treatment, cluster episodes were counted as a single seizure unless individual episodes were clearly documented. Two cats routinely had cluster episodes prior to levetiracetam treatment, and one of these cats did not have any additional cluster episodes after treatment with levetiracetam was begun. The other still had cluster episodes, although less frequently. Importantly, none of the cats reportedly had status epilepticus after treatment with levetiracetam was begun.

Levetiracetam was administered as an adjunct to phenobarbital in the cats enrolled in the present study. Ideally, we would have maintained stable phenobarbital concentrations throughout the study to be sure that changes in seizure frequency were attributable to levetiracetam treatment and not to changes in phenobarbital concentration. However, in 1 cat, phenobarbital administration was discontinued by the owner, and in another cat, the phenobarbital dosage was decreased by 50% because of a lack of seizures for several months and a serum phenobarbital concentration in the high end of the reference range. In both of these cats, seizure frequency increased when the phenobarbital dosage was decreased and seizure frequency decreased when phenobarbital dosage was increased again. Unfortunately, it was not practical to monitor serum phenobarbital concentrations in all cats frequently enough to be certain that concentrations were stable.

Levetiracetam was well tolerated in the cats in the present study and was not associated with any clinically important adverse effects. However, levetiracetam administration was discontinued in 1 cat because of a lack of efficacy (reduction in seizure frequency was only 4%) and ataxia. The ataxia persisted after administration of levetiracetam was discontinued, and it could not be determined whether the ataxia was related to the anticonvulsant treatment or was an unrelated problem.

Cats were eligible for enrollment in the present study only if a presumptive diagnosis of idiopathic epilepsy, defined as recurrent seizures without any apparent cause, had been made. There are no specific tests for idiopathic epilepsy in cats, and it is a diagnosis of exclusion. Unlike the case in dogs, there is no established age range for cats with idiopathic epilepsy, and cats in the present study ranged from 3 months to 16 years old at the time of seizure onset. It is our impression, however, that cats with idiopathic epilepsy are generally older at the onset of seizures than are dogs with idiopathic epilepsy.

In 1 cat in the present study, levetiracetam was initially given IV (40 mg/kg [18.2 mg/lb], as a bolus, followed by a continuous-rate infusion of 3.66 mg/kg/h [1.66 mg/lb/h]), rather than PO, because of status epilepticus at the time of initial examination. Seizure activity ceased 2 hours after treatment with levetiracetam was begun, and oral administration was begun 2 days later. No adverse effects were noted during infusion of levetiracetam, although the cat was sedate and recovering from a prolonged episode of status epilepticus. Additional study is needed to determine the safety and efficacy of IV levetiracetam administration in cats.

In people, food does not affect the extent of absorption of levetiracetam, although it delays the rate of absorption.^15,24 Therefore, it seems likely that levetiracetam can be safely administered independent of feeding schedule in cats, which eases the task of giving the drug 3 times daily. Renal elimination occurs primarily by glomerular filtration, and clearance correlates well with creatinine clearance.¹⁷ Therefore, the dosage should be decreased in patients with impaired renal function. However, levetiracetam does not undergo hepatic metabolism and should, therefore, be safe in cats with liver disease or impairment. Although the cost of levetiracetam is often prohibitive in canine patients due to size, this was not a factor in the feline population. Phenobarbital remains the first-line anticonvulsant in cats. However, when cats become refractory to or intolerant of phenobarbital, levetiracetam should be considered as an alternative anticonvulsant in cats with idiopathic epilepsy.