Incidence of hepatopathies in dogs administered zonisamide orally: A retrospective study of 384 cases

Abstract Background Acute hepatopathy secondary to administration of zonisamide has been reported in 2 dogs, but overall incidence of hepatopathy is unknown. Objective To characterize the incidence of hepatopathy in dogs administered zonisamide PO. Animals Three hundred eighty‐four dogs administered zonisamide PO. Methods Multicenter retrospective study. Medical records were searched for dogs prescribed zonisamide PO and which had follow‐up for at least 3 months (acute exposure) and >3 months (chronic exposure). Reported clinical signs, physical examination findings, and serum biochemical panels were reviewed for possible hepatotoxicosis. Serum alanine aminotransferase (ALT) and alkaline phosphatase (ALP) activity and albumin concentration were documented for all available cases. Results Acute clinical hepatopathy was found in 2 of 384 treated dogs (0.52%, 95% confidence interval [CI], 0.06‐1.9) after 13‐16 days of zonisamide treatment. One additional dog had elevated serum ALT activity with no clinical signs. Of these 3 dogs, 2 recovered after administration of zonisamide was stopped, and 1 was euthanized because of liver failure. Of the 117 cases chronically administered zonisamide, 10 had an increase in ALP, 6 had an increase in ALT, and 1 had hypoalbuminemia. No clinical signs of liver disease were noted in dogs chronically treated with zonisamide (median, 20 months; range, 5‐94 months). Conclusions and Clinical Importance Acute, potentially life‐threatening hepatopathy associated with oral administration of zonisamide to dogs is estimated to occur in less than 1% of dogs and was observed in the first 3 weeks of treatment. Subclinical abnormalities in ALT and ALP activity were noted in <10% of dogs during chronic administration of zonisamide, with no clinical signs of liver disease noted.


| INTRODUCTION
Zonisamide is considered a newer antiepileptic drug (AED) and is becoming commonly used for treating seizures in dogs. Zonisamide is currently labeled for adjunct treatment of partial seizures in humans. 1 In dogs with epilepsy, zonisamide is used for both monotherapy and adjunct management of seizures. 1 Limited studies in dogs have reported 60% efficacy when used as monotherapy and 58% to 80% as adjunct treatment. [2][3][4] Zonisamide is generally well-tolerated in dogs, and the most common adverse effects include decreased appetite, vomiting, sedation, and ataxia. 1 Idiosyncratic drug reactions occur, including neutropenia, erythema multiforme, and acute hepatopathy. 1 Specifically, 2 cases of severe, acute, idiosyncratic hepatopathy are reported in dogs. 5,6 In addition, persistent increases in serum alkaline phosphatase (ALP) activities and decreases in serum albumin concentration can occur in dogs treated experimentally with zonisamide at 5 times the recommended clinical dose (dosages used were 75 mg/kg/day). 7 However, the incidence of hepatopathies has not been evaluated in dogs treated clinically acutely or chronically with zonisamide.
The purpose of this retrospective study was to characterize the incidence of a hepatopathy associated with acute and chronic zonisamide administration in dogs with a history of seizures.

| Study cohort
The study design was a multicenter retrospective study. The medical records for dogs presented to Sage Veterinary Centers (Redwood City, Campbell, Dublin, and Concord, CA) and administered zonisamide PO from 2011 to 2018 were reviewed.

| Hepatopathy associated with acute zonisamide administration
For evaluation of the incidence of hepatopathy associated with acute zonisamide administration in dogs, inclusion criteria included dogs of any age, any epilepsy diagnosis, and at least 3 months of clinical follow-up after starting zonisamide. Dogs could be administered other medications at the time of first administration of zonisamide, but baseline bloodwork was required before oral administration of zonisamide, and dosing of other drugs had to remain unchanged during the 3-month observation period.
Adverse clinical signs noted during the 3-month period were abstracted from the medical record, including anorexia, hyporexia, vomiting, diarrhea, excessive lethargy, or icterus. Dogs were excluded if they had been diagnosed hepatopathy or concurrent systemic illness that could confound these clinical signs. Serum biochemical panels, if performed during this 3-month period, were reviewed for any abnormalities in alanine transferase (ALT) or ALP activity or albumin concentration. In order to maintain an accurate "denominator" for the number of treated dogs, performance of bloodwork was not required if clinical records indicated a clinically well dog at that visit.

| Hepatopathy associated with chronic administration of zonisamide
To evaluate for evidence of hepatopathies associated with long-term zonisamide administration, dogs receiving zonisamide PO for longer than 3 months were also included. For inclusion, follow-up blood work at any time point after first administration of zonisamide, including ALT and ALP activity and albumin concentration, was required. If any dog had an increased ALT or ALP activity, or decreased albumin concentration, documentation of normal values before starting zonisamide was required. Dogs concurrently receiving prednisone or phenobarbital were excluded, because of the induction of liver enzyme activity by these drugs. [8][9][10][11] Additionally, any dog diagnosed with a pre-existing hepatopathy or other systemic illness was excluded.

| Statistical analysis
Descriptive data are reported as percentages with 95% confidence intervals (CIs) for acutely treated dogs. The incidence of possible hepatopathy was expressed as a percentage by dividing the total number of cases with clinical or biochemical signs of acute hepatopathy by the total number of treated dogs that fit the inclusion criteria. For chronically treated dogs, the incidence of increases in ALT activity, increases in ALP activity, or decreases in albumin concentration was also calculated as percentages with 95% CI, using the total number of treated dogs with any biochemical screening as the denominator.

| Study cohort
After searching the electronic medical records system for dogs prescribed zonisamide at our hospital sites, 788 cases were identified and reviewed, and 384 dogs met the inclusion criteria for acutely treated dogs. The mean age was 6.2 years (range, 6 months to 17 years, SD 4.1) and mean weight was 19.1 kg (range, 1.6-63.6 kg, SD 13.1). For chronically treated dogs, 117 dogs met the inclusion criteria. The mean age was 4.6 years (range, 8 months to 16 years, SD 3.2) and mean weight was 18.4 kg (range, 2.3-74.2 kg, SD 13.9).

| DISCUSSION
This study investigates the incidence of hepatopathies administered zonisamide PO in pet dogs. An acute, clinical hepatopathy is reported in 2 isolated case reports in the veterinary literature. We found the incidence of acute, clinical hepatopathy to be rare, with an incidence of 0.57% in dogs receiving zonisamide administered PO. The detected overall incidence of acute biochemical hepatopathy was 2.9%; however, this is an estimate only because not all dogs had consistent biochemical monitoring. Although none of the 3 affected dogs underwent liver biopsies, the chronology of the increases in liver enzyme activity, along with resolution after discontinuation of zonisamide in 2 surviving dogs, are consistent with an acute adverse drug reaction to zonisamide. Given the low incidence and the time to onset in our study (13-16 days) and in 2 previous case reports (10-21 days), this is likely an idiosyncratic reaction in dogs. One fatal case report, along with the fatal outcome of case 2 in our study, suggests that this can be a life-threatening drug reaction, although recovery is possible for dogs with a milder clinical presentation.
In addition, dog 2 had been administered levetiracetam at the same time as zonisamide, and although there are no reports of a hepatopathy associated with levetiracetam in the veterinary or human medical literature, and acute idiosyncratic reaction cannot be ruled-out.
One dog had abnormal serum ALT activity 15 days after administering zonisamide with no accompanying clinical signs, and ALT activity normalized after discontinuation of zonisamide, suggesting the possibility of an acute biochemical hepatopathy. This dog highlights the potential value of rechecking serum liver enzyme activity in the first few weeks after initiating oral zonisamide treatment in dogs even if clinical signs are not present.
We also estimated the incidence of clinical or biochemical hepatopathies in pet dogs treated chronically with zonisamide for at least 5 months.
Observed increases in serum ALT (5.1% of treated dogs) and ALP (8.6% of treated dogs) activity were generally mild, although not all dogs had repeated biochemical monitoring. Only 1 dog developed mild hypoalbuminemia (0.85% of treated dogs) and no dogs developed hyperbilirubinemia.
Importantly, none of the dogs treated chronically with zonisamide developed any clinical signs associated these biochemical changes.
Our study suggests that increases in serum activity of hepatic enzymes (ALP and ALT) can occur during chronically administration of zonisamide, but at an apparently low incidence and without the development of clinical signs, even with continuation of zonisamide. Unfortunately, the majority of dogs with increased liver enzyme activity during chronically administration of zonisamide did not have further diagnostics performed, such as abdominal ultrasound, urinalysis, or bile acids. Therefore, the biochemical changes detected could have been because of another underlying etiology and unrelated to chronically administered zonisamide. Consistent with our findings, increases in serum ALT or ALP activity are found in 2% to 4% of people chronically administered zonisamide, with no clinically relevant hepatotoxicity reported. 12 There are several limitations to the current study related to the retrospective study design. In both cases of acute, clinical hepatopathy, acute hepatocellular injury with possible necrosis was suspected but was not histopathologically confirmed. Not all treated dogs had biochemical testing performed during the first 3 months of zonisamide administration.
In the chronically treated group, biochemical screening was inconsistent, measurement of bile acids was not performed, and some dogs were lost to follow-up. Additionally, it is possible that some dogs in the chronic administration group had acute, albeit subclinical, biochemical changes that were missed if bloodwork was not performed. For future prospective studies, consistent measurement of liver enzyme activity is needed to clarify the true incidence of biochemical hepatopathy in dogs treated with zonisamide and to define the number needed to test to determine the value of routine biochemical monitoring in these dogs.
In conclusion, our study found an apparent incidence of 0.52% for acute, likely idiosyncratic clinical hepatopathy during administration of zonisamide PO in 384 dogs. Because of the potentially life-threatening nature of acute hepatopathy, careful clinical monitoring and follow-up biochemical testing is essential during the first month of treatment with zonisamide. We found no evidence of clinically relevant hepatopathies in dogs treated with zonisamide for 5 months or longer.

ACKNOWLEDGMENT
No funding was received for this study.

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.