A retrospective study of the efficacy of zonisamide in controlling seizures in 57 cats

Abstract

Background

Evidence‐based recommendations for antiepileptic drug selection in cats beyond phenobarbital are limited, and additional studies are needed for cats where seizures remain inadequately controlled by administration of phenobarbital alone or for cats that cannot safely receive phenobarbital.

Objective

To compare seizure frequency in cats before and after oral administration of zonisamide and describe adverse clinical or clinicopathologic effects in this cohort.

Animals

Fifty‐seven cats with a history of seizures.

Methods

Multicenter, retrospective study. Median number of seizures per month and number of seizure days per month were compared before and after administration of zonisamide in all cats, a subgroup of cats with idiopathic epilepsy (IE), and a subgroup of cats receiving zonisamide as sole therapy. Clinical and clinicopathologic adverse effect data were also reported.

Results

A median decrease of 1 (P = .001, 95% confidence interval (CI) [−1.0, −0.5]) seizure per month, and 1 (P = .003, 95% CI [−1.5, −0.2]) seizure days per month was found across all cats after oral administration of zonisamide. The subgroup with IE showed median decreases of 1 (P = .03, 95% CI [−2.0, −0.5]) and 2 (P = .01, 95% CI [−2.5, −1.0]), respectively. The most common clinical adverse effects were sedation (17%), ataxia (11%), hyporexia (17%), and emesis (5%). One cat developed mild nonregenerative anemia, 2 cats developed mild metabolic acidosis, and 6 cats showed mild increases in ALT and ALP.

Conclusion

Zonisamide was well tolerated and efficacious in controlling seizure activity in most cats.

Abbreviations

AED

antiepileptic drug

CSF

cerebrospinal fluid

IE

idiopathic epilepsy

MRI

magnetic resonance imaging

1. INTRODUCTION

Epilepsy in cats is a common problem affecting an estimated 1% to 2% of the general domestic cat population, ¹ , ² and antiepileptic drugs (AEDs) are the mainstay of treatment. While phenobarbital is a safe and effective first‐line antiseizure medication for most cats, some cats cannot safely tolerate phenobarbital or have seizures that are not adequately controlled with phenobarbital alone. ³ , ⁴ , ⁵ In dogs, potassium bromide is a safe and effective alternative or add‐on AED; however, in cats, adverse reactions, including allergic‐like pneumonitis, occur in up to 40% to 50% making it no longer recommended. ⁶ Efficacy studies for AEDs beyond phenobarbital and potassium bromide are lacking in cats, and, unlike dogs, no consensus exists for seizure management. ³ , ⁷

Zonisamide is a newer AED approved in the United States and Europe as an add‐on AED to treat focal seizures in human adults. ² While used anecdotally for controlling seizures in cats, no efficacy studies evaluating zonisamide for seizures in cats have been published. Zonisamide is effective in animal models for controlling focal and generalized seizures, and clinical studies in dogs receiving zonisamide have shown 60% efficacy as a monotherapy and 60‐80% efficacy as an add‐on therapy. Existing studies on zonisamide in cats are limited. ⁸ , ⁹ , ¹⁰ While zonisamide appears safe and well‐tolerated in cats clinically, the efficacy of zonisamide for seizure control and the prevalence of adverse effects have not been established in cats with seizures. ³ , ¹¹

The primary goal of this study was to retrospectively identify epileptic cats receiving zonisamide as part of their seizure management and to evaluate monthly seizure frequency before and after starting zonisamide. We hypothesize that epileptic cats treated with zonisamide will have a significant decrease in seizure frequency. The secondary aim of this report is to perform a subgroup analysis of seizure frequency in cats diagnosed with idiopathic epilepsy. An additional secondary aim is to describe clinical and clinicopathologic adverse effects of cats on zonisamide therapy.

2. MATERIALS AND METHODS

2.1. Study sample

Medical records systems from the University of Wisconsin—Madison Veterinary Medical Teaching Hospital and SAGE Veterinary Centers (Redwood City, Dublin, Campbell, and Concord, CA) were searched retrospectively by species (domestic cat) and charge code (zonisamide 25 mg capsules, zonisamide 50 mg capsules) from January 2010 to December 2020, to capture cats that were prescribed zonisamide under the care of a board‐certified veterinary neurologist. Inclusion criteria were cats of any age, sex, and weight that presented for recurrent seizures (ie, >1 seizure episode), were prescribed zonisamide, and had a minimum follow‐up of 3 weeks. Criteria for exclusion from the study included incomplete records regarding seizure frequency, lack of clarity in terms of zonisamide dosing, and humane euthanasia within 3 weeks of seizure onset, as patterns and response to therapy were limited in these cases. Cases with partially incomplete follow‐up information were marked, and clients were contacted for any available follow‐up information to complete the data set. If additional and sufficient follow‐up was available, then these cases were included.

Demographic data collected for each case included age at the time of seizure onset, sex, breed, weight, definitive diagnosis for seizures (if available), all diagnostics performed, age at last follow‐up, and cause of death (if applicable). Clinicopathologic and diagnostic imaging information were reviewed when available, including complete blood counts and serum biochemistry panels before and after starting zonisamide, thoracic and abdominal radiography, abdominal ultrasound, magnetic resonance imaging (MRI), cerebrospinal fluid (CSF) analysis, infectious disease testing, and histopathology reports from surgical biopsies. Cases with no identifiable extracranial cause for seizures, MRI studies of the brain were no abnormalities were detected, and normal CSF analyses were classified as confirmed idiopathic epileptics for subgroup analysis.

2.2. Medication data

Medication data collected for each case included zonisamide start date, zonisamide dose and frequency, adverse effects reported by clients, follow‐up bloodwork, and dates and doses of any concurrently administered AEDs. Adverse effect data were recorded after starting zonisamide, including any evidence of hyporexia/anorexia, vomiting, sedation, or ataxia.

2.3. Seizure data

Seizure data collected included seizure phenotype (generalized or focal), whether cluster seizures were noted, and the average monthly seizure frequency before and after starting zonisamide. Seizure frequency was recorded as number of seizure events per month and number of seizure days per month to account for cluster seizure events. Details on the number of seizures were extracted from client‐reported histories in medical records and client‐provided seizure logs when available. When a range was provided for total seizures and seizure days per month, the midpoint was used as the record summary. The longest available follow‐up information was used when collecting seizure frequency data after starting zonisamide, for an average follow‐up period of 1.3 years.

2.4. Outcome measures

The primary outcome measure was the change in monthly seizure frequency and monthly seizure days by magnitude and percentage before and after starting zonisamide for all cats and then in a subgroup of cats diagnosed with idiopathic epilepsy. The percentage of cats that responded to zonisamide was calculated for the total study sample and for the subgroup of cats diagnosed with idiopathic epilepsy. Response to therapy was defined as >50% reduction in seizure frequency. The secondary outcome measure was the frequency of clinical and clinicopathologic adverse effects after starting zonisamide. Another analysis was done comparing cats that were on other AEDs and those receiving zonisamide as sole therapy.

2.5. Statistical analysis

Data were assessed for normality by a Shapiro‐Wilk test. Demographic data were analyzed with descriptive statistics, and medians and ranges were reported for nonnormal data. Confidence intervals (CIs) around response to therapy percentages are created by Wilson's score method. Seizure frequency data were evaluated for normality, and median values were compared using nonparametric statistical tests for nonnormal data. Differences in seizure frequency and seizure days between the prezonisamide and postzonisamide treatment groups were assessed by a paired, two‐sided Wilcoxon signed‐rank test and a nonparametric paired randomization test. A bootstrap sampling distribution was created by resampling the same size dataset 1000 times with replacement by patient from the entire dataset and calculating the median difference. Nonparametrically bootstrapped 95% confidence intervals are created by taking the 2.5th, and 97.5th percentile of the bootstrap sampling distribution. Similar testing and confidence interval testing was done within subgroups of cats with zonisamide used as an add‐on therapy and cats with zonisamide used alone. The seizures and seizure days outcomes were tested and bootstrapped separately. Statistical significance was defined as P < .05. The association of zonisamide dosage frequency with differences in seizure frequency and seizure days were evaluated by Wilcoxon rank‐sum tests. The association of zonisamide dosage frequency with adverse effects was tested by a Cochran‐Mantel‐Haenszel (CMH) test without continuity correction. Overall frequency of adverse effects is reported as a common odds ratio between q24h and q12h dosing. Similar CMH testing was done for association of other AED usage and adverse effects. All statistical analysis was performed in the statistical computing environment R, version 4.1.2 (https://www.R-project.org/).

3. RESULTS

3.1. Study sample and subgroup characteristics

The initial medical record search yielded 103 cases, 46 of which were excluded due to incomplete records on seizure frequency (29), lack of clarity in terms of zonisamide dosing (11), humane euthanasia within 3 weeks of seizure onset (4), or never starting zonisamide (2). Therefore, a total of 57 cats were included in the final analysis.

The study sample consisted of 33 spayed females, 2 intact females, 21 castrated male cats, and 1 intact male cat (Table 1). The median age of seizure onset for the study sample was 7 years (range, 0.17‐22 years), and the median weight was 4.5 kg (range, 2.1‐8.1 kg). Breeds represented included 32 domestic shorthair cats, 6 domestic longhair cats, 3 domestic medium hair cats, and 16 purebred cats (American Shorthair [n = 2], Bengal [n = 3], Birman [n = 1], Maine Coon [n = 3], Ragdoll [n = 2], Russian Blue [n = 1], Savannah [n = 1], Siamese [n = 2], and Turkish Van [n = 1]).

TABLE 1.

Study sample data for cats administered zonisamide orally for seizures.

		All cats (n = 57) a	Confirmed idiopathic subgroup (n = 16) a
Age at 1st seizure (years)		7 (0.17, 22.00)	8 (0.25, 20.00)
Age at last follow‐up (years)		11 (0.50, 22.00)	10 (0.50, 22.00)
Weight (kg)		4.5 (2.10, 8.10)	4.64 (2.30, 8.10)
Sex	Female intact	2 (4%)	1 (6%)
	Female spayed	33 (58%)	7 (44%)
	Male intact	1 (2%)	0 (0%)
	Male castrated	21 (37%)	8 (50%)
Seizure phenotype	Generalized	29 (51%)	7 (44%)
	Focal	12 (21%)	2 (13%)
	Both	15 (26%)	7 (44%)
	Unknown	1 (2%)	0 (0%)
Concurrent AEDs (other than zonisamide)	Yes	33 (58%)	9 (56%)
	No	24 (42%)	7 (44%)

Note: Antiepileptic drugs (AEDs).

^a Table entries are in format “median (minimum, maximum)” and “count (% of population).”

The median age of seizure onset in our study sample was 7.0 years (range, 0.17‐22.0) and the median age at the initiation of zonisamide therapy was 9.0 years (range, 0.25‐22.0). The median age at last follow‐up was 11 years (range, 0.5‐22.0 years), and at the time of data collection, 16 cats were no longer alive. The cause of death was humane euthanasia in 8 cats and unknown in 8 cats. Reasons for humane euthanasia included progressive seizure activity in 3 cats, progressive intestinal disease in 1 cat, pulmonary disease in 1 cat, urinary obstruction in 1 cat, and was not specified in 2 cats. The average time period from the first noted seizure until the time of starting oral zonisamide therapy was 6 weeks.

Diagnosis via advanced imaging was available in 27 of 57 cats (47%). Idiopathic epilepsy, defined as an MRI and CSF tap where no abnormalities were detected, was confirmed in 16 of 27 cats (59%). Neoplasia was confirmed on histopathology following surgical resection in 4 cats, which revealed meningioma in 3 cases and lymphoma in 1 case. Congenital brain abnormalities were diagnosed in 4/27 (15%) cases (hydrocephalus [n = 2], intracranial cyst [n = 1], unspecified congenital brain abnormality [n = 1]). Two cats had a history of head trauma, and 1 cat was diagnosed with metabolic encephalopathy. The remaining 30 (of 57 total) cats did not have advanced diagnostics, including imaging, performed.

For the subgroup of cats with confirmed idiopathic epilepsy (16 cats), the median age of seizure onset was 8 years (range, 0.25‐20 years) and the median weight was 4.64 kg (range, 2.3‐8.1 kg). Breeds represented included 8 domestic shorthair cats, 2 domestic longhair cat, 2 Maine Coons, 1 domestic medium hair cat, 1 American shorthair cat, 1 Birman, and 1 Ragdoll.

3.2. Seizure data—Entire cohort

Before initiating zonisamide therapy, seizures were generalized only in 29/57 cats, focal only in 12/57 cats, and both focal and generalized in the remaining 15/57 cats. One record did not specify generalized versus focal seizures. Most cats (39/57) had episodes of cluster seizures. The median values and ranges for the number of seizures per month and seizure days per month are summarized in Table 2. The number of seizures per month was significantly decreased by a median of 1 seizure after starting zonisamide (P = .001, 95% CI [−1.0, −0.5], Figure 1). The number of seizure days per month was significantly decreased by a median of 1 day after oral administration of zonisamide (P = .003, 95% CI [−1.5, −0.2], Figure 1). The percentage of cats that responded to zonisamide considering number of seizures per month was 62%, or 28 out of 45 cats with efficacy data available (95% CI [48%, 75%]).

TABLE 2.

Association of zonisamide administration on monthly seizure frequency in study cohort and in the subgroup of cats with confirmed idiopathic epilepsy.

	Before zonisamide a	After zonisamide a	Wilcox P‐value	n	Median difference b	% responders c
Overall
Seizure days	2 (0.0, 21.0)	0.5 (0.0, 15.5)	P = .003	43	−1.0 [−1.5, −0.2]	65% [50, 78]
Seizure episodes	1 (0.5, 17.0)	0.5 (0.0, 16.5)	P = .001	45	−1.0 [−1.0, −0.5]	62% [48, 75]
Confirmed idiopathic
Seizure days	3 (0.5, 21.0)	0.8 (0.0, 5.5)	P = .01	13	−2.0 [−2.5, −1.0]	77% [50, 92]
Seizure episodes	2 (0.5, 9.0)	1 (0.0, 5.5)	P = .03	13	−1.0 [−2.0, −0.5]	69% [42, 87]

Note: Responders are cats with greater than 50% reduction in seizure frequency.

^a Table entries are in format “median (minimum, maximum).”

^b 95% bootstrap confidence intervals (CI) are around the median difference of postminus prezonisamide treatment within cat.

^c 95% CIs are calculated by Wilson's score method.

FIGURE 1.

Monthly seizure frequencies before and after oral zonisamide administration in the A) overall study sample and B) confirmed idiopathic epilepsy subgroup. Lines trace individual cats from before to after oral zonisamide administration. Marginal kernal density estimates are displayed along both axes with the median highlighted.

3.3. Seizure data—Confirmed idiopathic epilepsy subgroup

In the idiopathic epileptic subgroup, seizure phenotype was generalized only in 6/16 cats, focal only in 3/16 cats, and generalized and focal in the remaining 7/16 cats. The number of seizures per month and seizure days per month before and after oral administration of zonisamide are summarized in Table 2. The number of seizures per month was significantly decreased by a median of 1 seizure (P = .03, 95% CI [−2.0, −0.5], Figure 1). The number of seizure days per month was significantly decreased by a median of 2.0 days (P = .01, 95% CI [−2.5, −1.0], Figure 1). The percentage of cats with idiopathic epilepsy that responded to zonisamide considering number of seizures per month was 69.2%, or 9 out of 13 cats with efficacy data available (95% CI [42%, 87%]).

3.4. Seizure data—Zonisamide as add‐on therapy

Thirty‐three cats were already receiving another AED when zonisamide treatment was initiated. The demographics are shown in Table 3. In cats receiving zonisamide as add‐on therapy, the number of seizures per month was significantly decreased by a median of 0.5 seizure (P = .05, 95% CI [−1.3, 0.0], Figure 2). The number of seizure days per month was significantly decreased by a median of 1.0 days (P = .02, 95% CI [−2.0, 0.0], Figure 2). The percentage of cats receiving zonisamide as add‐on therapy that responded to zonisamide was 56% (95% CI [37%, 73%]).

TABLE 3.

Study sample and seizure frequency data for cats administered zonisamide alone or cats administered zonisamide concurrently with other AEDs.

		Zonisamide alone (n = 24) a	Concurrent with other AEDs (n = 33) a
Age at 1st seizure (years)		15.0 (1.5, 18.0)	4.5 (0.2, 22.0)
Weight (kg)		4.4 (2.1, 8.1)	4.5 (2.5, 6.6)
Idiopathic	Yes	7 (29%)	9 (27%)
	No	17 (71%)	24 (73%)
Sex	Female intact	0 (0%)	2 (6%)
	Female spayed	14 (58%)	19 (58%)
	Male intact	0 (0%)	1 (3%)
	Male castrated	10 (42%)	11 (33%)
Seizure type	Both	8 (33%)	7 (21%)
	Generalized	10 (42%)	19 (58%)
	Partial	6 (25%)	6 (18%)
	Missing	0 (0%)	1 (3%)

^a Table entries are in format “median (minimum, maximum)” and “count (% of population).”

FIGURE 2.

Monthly seizure frequencies before and after oral zonisamide administration split into subgroups based on whether zonisamide was used concurrently with other anti‐epileptic drugs. Lines trace individual cats from before to after oral zonisamide administation. Marginal kernal density estimates are displayed along both axes with the median highlighted.

3.5. Seizure data—Zonisamide as sole therapy

Twenty‐four cats received zonisamide as a sole AED treatment, and the demographics of this study sample are shown in Table 3. In cats receiving zonisamide as sole therapy, the number of seizures per month was significantly decreased by a median of 1 seizure (P = .001, 95% CI [−1.5, −0.5], Figure 2). The number of seizure days per month was significantly decreased by a median of 1 day (P = .0004, 95% CI [−2.0, 0.0], Figure 2). The percentage of cats receiving zonisamide as sole therapy that responded to zonisamide was 70% (95% CI [48%, 86%]). Full results of this cohort are shown in Table 4.

TABLE 4.

Association of zonisamide administration on monthly seizure frequency in subgroups of cats with zonisamide as the sole therapy, and cats using zonisamide with other AEDs.

	Before zonisamide a	After zonisamide a	Wilcox P‐value	n	Median difference b	% responders c
Alone
Seizure days	2 (0.0, 16.0)	0.1 (0.0, 3.0)	P = <.001	17	−1.0 [−2.0, 0.0]	77% [53, 90]
Seizure episodes	1 (0.5, 16.5)	0.5 (0.0, 3.0)	P = .001	20	−1.0 [−1.5, −0.5]	70% [48, 86]
With other AED
Seizure days	2.2 (0.0, 21.0)	0.5 (0.0, 15.5)	P = .02	26	−1.0 [−2.0, 0.0]	58% [39, 75]
Seizure episodes	1 (0.5, 17.0)	1 (0.0, 16.5)	P = .05	25	−0.5 [−1.3, 0.0]	56% [37, 73]

Note: Responders are cats with greater than 50% reduction in seizure frequency.

^a Table entries are in format “median (minimum, maximum).”

^b 95% bootstrap confidence intervals (CI) are around the median difference of after minus before zonisamide treatment within cat.

^c 95% CIs are calculated by Wilson's score method.

3.6. Medication data

The median daily dose of zonisamide administered to the study sample was 7.55 mg/kg (range, 3.8‐17.7 mg/kg). The dosage frequency was q24h in 21/57 (37%) cats and q12h in 36/57 (63%) cats. The average zonisamide dose in the subgroup of cats with idiopathic epilepsy was 6.90 mg/kg (range, 4.4‐17.7 mg/kg). The dosage frequency in this subgroup was q24h in 1/16 and q12h in 15/16 cats. A dosage frequency of q12h is estimated to improve seizure control (difference of pre‐ and postzonisamide seizure clusters) by 0.5 seizures compared with q24h, but the association is not statistically significant (95% CI [−2.0, 0.5], P = .18).

Additional drugs were used in the maintenance AED regimen for 33/57 (58%) cats, (phenobarbital only [n = 20], levetiracetam only [n = 5], and both phenobarbital and levetiracetam [n = 8]). In the idiopathic epileptic subgroup, 8/16 cats were concurrently receiving phenobarbital and/or levetiracetam (phenobarbital only [n = 6], levetiracetam only [n = 1], and both [n = 2]).

A total of (15/57) cats experienced at least 1 adverse effect after starting zonisamide (that were not present beforehand). Adverse effects reported included inappetence (n = 10, duration 1.5 days to 4 weeks, for 1 cat it persisted until zonisamide was discontinued), transient sedation (n = 6, duration 2‐4 weeks), ataxia (n = 4, duration 2‐4 weeks), and vomiting (n = 3, duration 1 day to 2 weeks). A trend of increased adverse effects with increased dosage frequency of zonisamide (q12h vs q24h) was observed with weak statistical significance (P = .04). The q12h dosage is estimated to have 3.2 times higher odds of reporting adverse effects (95% CI [1.02, 9.85]) than the q24h dosage across adverse effects. There was no significant association of reported adverse effects of zonisamide when used concurrently with other AEDs (OR = 0.81, 95% CI [0.34, 1.93], P = .63).

Complete blood counts (CBC) were available after starting zonisamide in 11 cats. No CBC abnormalities were identified in 6 cats, and a mild nonregenerative anemia was noted in 3 cats that was unchanged compared to their prezonisamide CBC data. One of these cats was receiving levetiracetam and amlodipine, and the other 2 cats were not receiving any other medications. One cat had a new mild nonregenerative vs preregenerative anemia (HCT 24.4%, reticulocytes 9900 cells/μL) not present on the prezonisamide CBC, and 1 cat had a new thrombocytosis and lymphopenia that was attributed to systemic disease (suspected gastrointestinal lymphoma). This cat was also receiving prednisolone. Serum biochemistry panels were available before and after starting zonisamide in 27 cats. Six cats had mild hepatocellular and cholestatic liver enzyme changes above the reference interval (postzonisamide liver value ranges: ALT range 151‐349 U/L, AST range 70‐131 U/L, ALKP 90‐308 U/L). Two cats had mild metabolic acidosis (TCO2 15 mmol/L [16‐25]; TCO2 10 mmol/L [12–22]).

4. DISCUSSION

This retrospective study supports the use of zonisamide as an apparently effective alternative or add‐on therapy for controlling seizures in cats, with an overall response rate of 56% to 77%, similar to zonisamide efficacy studies in dogs. ² Cats with seizures had a significant decrease in seizure frequency after oral administration of zonsamide compared to seizure frequency before starting zonisamide.

Antiepileptic drug efficacy studies in cats have historically been focused on phenobarbital with fewer newer studies on imepitoin and levetiracetam. ¹² , ¹³ , ¹⁴ Recent studies reviewing phenobarbital reactions in cats suggest that type II (unpredictable, dose‐independent) idiosyncratic reactions are rare. ³ , ⁴ , ⁵ Individual case reports on such reactions include bone marrow suppression, fever, and pseudolymphoma. Additional drug options shown to be effective are needed for this cohort of cats that cannot tolerate phenobarbital. In addition, alternative AED therapy is also needed by cats whose seizure frequency cannot be adequately controlled with phenobarbital alone, as well as cats that cannot receive phenobarbital due to liver dysfunction. Lastly, unlike dogs, adverse reactions to potassium bromide are common and well‐documented in cats. ⁶ Potassium bromide is not recommended in cats due to risk for pneumonitis in up to 40% of cases, which may be fatal. ³ , ¹⁵ Therefore, a true need for further efficacy studies of other AEDs in cats with seizures is highlighted.

Zonisamide has multiple antiseizure mechanisms of action, including inhibition of voltage‐gated sodium and T‐type calcium channels and neuromodulatory effects facilitating inhibitory neurotransmitter signaling. ² Other actions of zonisamide include free radical scavenging, which may be play a role in neuroprotection, and weak inhibition of the enzyme carbonic anhydrase. The multiple mechanisms of action of zonisamide may improve seizure control to varying extents in different subsets of cats with seizures. Etiologies for epilepsy in cats include idiopathic epilepsy, structural (symptomatic) epilepsy, and reactive seizures. ¹⁶ , ¹⁷ Similarly, seizure manifestations in cats are highly variable, ranging from difficult‐to‐recognize focal seizures to classic generalized tonic‐clonic seizures, with various reports of different types of complex focal seizures throughout the literature (ie, audiogenic reflex seizures, complex focal seizures with orofacial and autonomic involvement). ¹⁸ , ¹⁹ , ²⁰ The multiple mechanisms of action of zonisamide may improve seizure control to varying extents in different subsets of the feline epilepsy population. Further studies are needed to clarify the potential of zonisamide in controlling seizure activity in different subsets of feline epileptics, both in terms of etiology and seizure semiology.

Few cats in this study sample had an increase in seizure frequency after administration of zonisamide. This finding was expected, as a proportion of cats with seizures have drug‐resistant epilepsy. Statistically significant decreases in seizure frequency overall were identified despite this characteristic of the data set. Additionally, only 56% of cats that received zonisamide as add‐on AED therapy responded, but this subgroup would be considered refractory (ie, needing more than 1 medication).

The low incidence of adverse effects in the study is consistent with previous reports, including sedation, ataxia, and gastrointestinal upset. ³ However, the duration of these adverse effects was longer than expected (up to 2‐4 weeks). The half‐life of zonisamide in cats is approximately 33 hours, such that the drug should reach steady state in approximately 1 week. ⁸ Clinicians should be aware that sedation, ataxia, and gastrointestinal upset may take several weeks to resolve after starting zonisamide in cats. Our analyses showed a trend for increasing adverse effects with q12h dosage than q24h dosage that is weakly significant without a strong effect of further reducing seizure frequency, suggesting clinicians should begin zonisamide with q24h dosing.

Clinicopathologic changes noted after oral administration of zonisamide in this study sample included mild increases hepatocellular and cholestatic liver enzymes in 6 cats and a mild metabolic acidosis in 2 cats. The mild metabolic acidosis noted in 2 cats is likely explained by the carbonic anhydrase inhibition properties of zonisamide. ²¹ Lastly, no cats in this study developed anticonvulsant hypersensitivity syndrome, which has been reported in 1 cat receiving zonisamide. ¹¹ Overall, the low frequency of clinicopathologic abnormalities in this study sample supports the safety profile of zonisamide in cats, although larger cohorts are needed for the true prevalence of adverse effects and not every cat had follow‐up bloodwork performed.

A major limitation of this retrospective study is the reliance on client reported seizure frequency documented in medical records. Client‐reported seizure frequency is subjective and may be unreliable, and one retrospective study in dogs comparing client‐reported seizure frequency to ictal electroencephalography data suggested that the true seizure frequency may be underreported by clients. ²² Bias in seizure frequency reporting is compounded by the lack of standardization in which owners were asked about seizure frequency; thus, it could be inflated if they are recalling acute seizure exacerbations, and could be deflated since most owners are not able to directly supervise their cats 24 hours a day. The period of time the cats are evaluated for seizure frequencies is inconsistent between clients due to the retrospective nature of the study. Furthermore, the placebo effect has been shown to reduce client‐reported seizure frequency in canine epilepsy studies. ²³ The reliance on client‐reported seizure frequency will likely continue to limit clinical veterinary epilepsy research until the use of electroencephalography becomes more commonly used in veterinary medicine.

Other limitations of this study include the variation in concurrently used AEDs in a majority of cats, such as phenobarbital and levetiracetam, and the variation in dose and frequency of zonisamide used between cases. Concurrent use of other AEDs would potentially impact seizure frequency but reflects difficulty in managing seizures in individual patients. However, by separating out these groups, we have hopefully been able to document the effect of oral administration of zonisamide as a sole AED. Likewise, the varying dose and the frequency of zonisamide may have also affected the seizure frequency in this study sample. Lastly, the response to zonisamide may differ in cats with different seizure etiologies, and this study was not powered enough to analyze response to zonisamide in cats with confirmed diagnoses beyond idiopathic epilepsy. Another possible bias includes a variable run‐in time for prescribing zonisamide to a patient since their first seizure, especially in cases where other AED(s) were being used. Information about when previous AED's were stopped, or adherence to treatment plans was also incomplete in medical records, which may introduce further variability in this study.

However, these variabilities reflect the clinical reality of managing seizure patients, and significant differences in data before and after starting zonisamide were obtained despite these limitations. Results should be interpreted within the scope of the study sample, and randomized controlled clinical trials are needed for stronger generalizability. Further prospective clinical trials are warranted to clarify the most appropriate dose and frequency of zonisamide, as well as to determine the effectiveness of zonisamide as a monotherapy and as an adjunct therapy to phenobarbital.

5. CONCLUSION

Monthly seizure frequency was significantly decreased after administering zonisamide, with an acceptable low level of adverse effects in this retrospective evaluation of epileptic cats. Adverse effects, such as sedation, ataxia, and gastrointestinal upset, were mostly transient but could take up to several weeks to resolve. Zonisamide can be used as an alternative or an adjunct therapy to phenobarbital or levetiracetam for seizure control in cats. Additional prospective studies are needed to determine the most appropriate dose and frequency in different samples of cats with epilepsy.

CONFLICT OF INTEREST DECLARATION

Authors declare no conflict of interest.

OFF‐LABEL ANTIMICROBIAL DECLARATION

Authors declare no off‐label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION

Authors declare no IACUC or other approval was needed.

HUMAN ETHICS APPROVAL DECLARATION

Authors declare human ethics approval was not needed for this study.

Djani DM, Liou M, Aravamuthan S, Lau V, Cameron S. A retrospective study of the efficacy of zonisamide in controlling seizures in 57 cats. J Vet Intern Med. 2024;38(2):1092‐1100. doi: 10.1111/jvim.16984.