The liver is the main detoxifying organ in the body. Continuous exposure to drugs, tox- ins, and different infections can trigger liver injury and eventually lead to various liver diseases (Stephens et al., 2014). Susceptibil- ity of the liver to injury is much higher than any other organ because it biotransforms most xenobiotics and receives blood directly from the gastrointestinal tract (Kumar et al., 2015). Phenobarbital is one of the most pop- ular anticonvulsant medications used in the veterinary field today. It is highly effective for the treatment of epilepsy in small ani- mals. Phenobarbital is primarily metabolized by the liver and induces the cytochrome P450 enzyme activity, which can lead to progres- sive shortening of the elimination  half-life of barbiturates with chronic administration (Munana et al., 2015).

The elimination half-life of phenobarbital is approximately 40 hours in cats, although sig- nificant variability may be noted among them. Typically,  the starting dosage for cats is 2 to  4 mg per kg of body weight (B.W.) This can be increased up to 16 mg per kg of B.W. per day based on the animal’s response (Podell, 2013).

Long-term administration of phenobarbi- tal has been reported to cause hepatic injury (hepatotoxicity) (Dayrell-Hart et al., 1991). Some dogs and cats receiving phenobarbital develop increases in certain liver enzymes, especially alkaline phosphatase (ALP) and aspartate transaminase (AST). Phenobarbital should be administered with extreme caution in pets diagnosed with hypovolemia, kidney disease, anemia, and hypoadrenocorticism. The use of the medication should be avoid- ed outright in pets with severe respiratory dysfunction, liver disease, or those with a known sensitivity to other anti-seizure med-

ications (Muller et al., 2000). Treatment of seizures in cats is similar to dogs, with only  a few limitations related to species-specific drug toxicities (Hazenfratz et al., 2018).

In many studies, silymarin has long been used in the treatment of liver disorders, in- cluding acute and chronic hepatitis, toxin/ drug-induced hepatitis, and other liver dis- eases (Lappin, 2001; Lo et al., 2014). There are few reported adverse effects of silymarin in cats. Beneficial effects on minimizing the formation of free radicals associated with reperfusion of ischemic have  been  report- ed (Lappin, 2001; Sun et al., 2018). The cat has been used extensively as an experimen- tal model for studying the pharmacology of drugs (Das and Vasudevan, 2006).

Cats are extremely sensitive to the toxic ef- fects of different drugs. These animals form glucuronides with many compounds slowly, or not at all because they possess fewer iso- forms of the enzymes that mediate the con- jugation. Cats have a relative deficiency of a specific high-affinity glucuronyl transferase that conjugates drugs with glucuronic acid. The deficiency of the glucuronide conjuga- tion pathway results in more drugs being conjugated to sulfates; however, the sulfation pathway has a finite capacity, which is also lower in cats than the other species (Lappin et al., 2001).

During prolonged treatment with different drugs, measurement of serum enzyme activ- ities is necessary. Clinical signs and labora- tory test results are nonspecific and do not differentiate drug-induced from other causes of hepatic diseases (Muller et al., 2000). El- evation in alanine aminotransferase (ALT) and AST activities are the most consistent findings. Serum ALP and lactate dehydroge- nase (LDH) activities also may be increased.

Bilirubinuria and hyperbilirubinemia occur more commonly in cats than in dogs (Kayne and Jepson, 2004).

Silymarin was chosen because of the anti- oxidant properties. This drug may protect the liver against oxidative stress through direct reactions with free radicals. Silymarin most likely inhibits TNF-α production in the liver (Gabrielova et al., 2015; Singh et al., 2016). Based on the aforementioned background, the present study was conducted to evaluate the possible hepatoprotective effects (thera- peutic and prophylactic) of silymarin against experimental phenobarbital-induced hepato- toxicity in cats.

Materials and methods

Animals

Twenty-four adult cats, DSH breed, 1-1.5 years old, and weighing 2.35-3.9 kg were randomly allotted to four equal groups. All cats appeared healthy, as determined by clinical examination, normal  hemogram,  and clinical biochemical profiles. The cats were kept in separate cages at 22-25˚C with 12-h light/12-h dark cycles, in a controlled environment. They were fed with a home- made diet containing chicken and fish. Wa- ter was provided ad libitum. This study was approved by the animal care program and research committee of Shahid Chamran Uni- versity of Ahvaz. It was conducted based on the guidelines for small animal care and use (code of ethics: 1397.57202). Cats in group A were given phenobarbital (DarouPakhsh Pharmaceutical Co., Tehran, Iran) with dosage 16 mg/kg (as a  toxic  dose)  orally for 28 days in gelatin capsules; group B re- ceived silymarin with dosage 30 mg/kg/day (Jalinous Pharmaceutical Co., Tehran, Iran) and continued for twenty-eight days orally concurrent with phenobarbital; groups C and

D were treated like group B, but silymarin was administered 3 and 48 h after adminis- tration of phenobarbital and continued up to twenty-eight days. Blood samples were col- lected from the jugular vein under general anesthesia (a combination of ketamine with dosage 15 mg/kg and acepromazine 0.15 mg/kg) for all groups. After collecting blood in centrifuge tubes, the tubes were  allowed to coagulate at room temperature and then placed in a water bath at 37˚C for 10 min. Centrifugation was performed at 1000 g for 20 min. The clear serum was separated and used for the analysis of biochemical parame- ters. The serum concentrations of ALP, ALT, AST, and LDH (as indices of liver injury), total and direct bilirubin, and kidney indices (BUN and creatinine) were measured before administration of phenobarbital and 24 h, 72 h and 28 days later. Serum concentrations of the above indices were measured in an au- tomated chemical analyzer (BT 3000 Plus, Biotechnica, Milan, Italy) using diagnostic kits (Pars Azmoon Co., Tehran, Iran).

Statistical analysis

The arithmetic mean of ALP, ALT, AST, LDH, bilirubin (total and direct), BUN and creatinine were compared among groups using repeated measures ANOVA, one way analysis of variance and LSD test (SPSS, version 10, SPSS Inc., Chicago, IL, USA). Differences were considered  significant when P≤ 0.05.

Results

In the present study, there were significant differences in enzyme activities among the different treatment groups (P <0.001). Ac- cording to the results presented in  Tables 1-4, phenobarbital elevated significantly se- rum concentrations of ALP, ALT, AST, LDH (in all cases), and total and direct bilirubin

in two cats of group A, after 24 h. Group, time, and the interaction of group and time had a significant effect  on changes  of liv-  er enzymes (P<0.001). In groups, B and C, levels of serum enzyme activities and total and direct bilirubin remained within the nor- mal range up to 28 days (P >0.05), but in the group D, levels of serum enzyme activities were higher than normal values (P<0.001) (in four cases) and total and direct bilirubin remained within the normal range. The vari- ables were different according to the drug administration time in group D. Prophylactic

(time zero) or therapeutic (3 h later) appli- cation of silymarin had a significantly ben- eficial effect and prevented the increase in serum enzyme activities, and total and direct bilirubin in both groups B and C. BUN (18- 27 mg/dl) and creatinine (0.9-1.8 mg/dl) lev- els were normal in all groups. Clinical find- ings mainly included depression, sedation, and ptyalism, during the study period. None of the cats died during the experiment. The course of the enzymatic activities is shown in Tables 1 to 4 (mean ± SD).

Table 1. The mean ± SD of serum ALP concentration (IU/L) in experimental groups of cats based on group and time

Significant differences are presented by capital letters in each row and lower case letters in each column (P<0.05).

Table 2. The mean ± SD of serum ALT concentration (IU/L) in experimental groups of cats based on group and time

Significant differences are presented by capital letters in each row and lower case letters in each column (P <0.05).

Table 3. The mean ± SD of serum AST concentration (IU/L) in experimental groups of cats based on group and time

Significant differences are presented by capital letters in each row and lower case letters in each column (P <0.05).

Table 4. The mean ± SD of serum LDH concentration (IU/L) in experimental groups of cats based on group and time

Significant differences are presented by capital letters in each row and lower case letters in each column (P<0.05).

Discussion

The results of the present study showed that silymarin had inhibitory effects on phe- nobarbital-induced hepatotoxicity in cats of groups B and C completely so that the pa- rameters remained within the normal range. In group D, relatively similar to group A, the variables were different according to the time of drug intake. In the present study, the oral administration of phenobarbital (16 mg/kg) for 28 days caused chronic hepatotoxicity in cats as verified by clinical and biochemical investigations. Administration of phenobar- bital increased serum concentrations of liver enzymes (in all cats) and total and direct bil- irubin (in two cats) of group A. The result of the therapeutic dose of silymarin (up to 3 h after phenobarbital administration) was sim- ilar to prophylactic dose (co-administration of both the drugs) in the prevention of hepa- totoxicity in our study, but administration of silymarin did not have preventative effects after 48 h.

Antioxidative and anti-inflammatory ef- fects of silymarin (by a significant decrease in hepatic myeloperoxidase activity and nitric oxide production) are known in rat models (Kumar et al., 2015). Silymarin can prevent liver injury by the inhibition of free radical formation. The recommended dosage for silymarin is 20-50 mg/kg/day in dogs and

cats (Hsu, 2013). It is reported that the ben- eficial effects of silymarin on phenobarbi- tal-induced liver injury may be attributed to the effect of the former on aldehyde oxidase mediated metabolism of  phenobarbital  in the liver (Singh et al., 2016). Phenobarbital likely increases the effects of the inhibitory neurotransmitter gamma aminobutyric acid (GABA), and decreases the release of the ex- citatory neurotransmitter glutamate  (Podell et al., 2016).

In the present  study,  we  used  silymarin for 28 days; nevertheless, the duration of ad- ministration depends on the condition being treated, response to the medication, and the development of any  adverse effects. This medication is available in a variety of formu- lations, including oral tablets and injectable forms. Dosage is determined based on several different factors, including the pet’s species, breed, size, and age. Phenobarbital hepatotox- icity can be managed with drug discontinua- tion or substantial dose reduction.

Based on the presumption that phenobar- bital is a hepatotoxic drug, an increase in serum enzyme activities within 24 h after onset, indicates a need to suspend drug ad- ministration and to provide supportive care (Khoury et al., 2015). Reports of hepatotox- icity associated with phenobarbital adminis- tration in veterinary medicine are exclusively limited to dogs; however, vigilance for these

adverse effects in cats is warranted, and rou- tine monitoring of serum chemistry profiles is recommended in feline patients receiving this drug. The most common adverse reactions of phenobarbital reported in dogs are sedation, weight gain, increased thirst (polydipsia), in- creased urination (polyuria), and increased appetite, most of which resolve within a few weeks of starting treatment. Less common but more serious adverse effects of  phenobarbi-  tal administration include acute idiosyncratic liver or bone marrow toxicity. Additional rare adverse effects of phenobarbital specifically reported in cats include facial pruritus, limb edema, thrombocytopenia, and cutaneous and mucosal hypersensitivity reactions (Hazenfra- tz et al., 2018). In the present study, the most common clinical signs were depression, seda- tion, and ptyalism. All cats completed the study without mortality. We did not see signs of hy- persensitivity reactions as observed in dogs.

The role of oxidative stress in phenobarbi- tal-induced hepatotoxicity and the therapeutic effects of natural antioxidants have been eval- uated in some studies. For example, Muller et al. (2000) suggested that increases in concen- trations of ALP, ALT, and GGT may reflect enzyme induction rather than hepatic injury during phenobarbital treatment in dogs.

In another study, serum ALP isoenzyme activities were determined for 23 epileptic dogs before the start of phenobarbital and at three weeks, six months, and 12 months af- ter the start of treatment. Median serum ALP was significantly increased at six months and 12 months compared with time zero (Gaskill et al., 2005).

Silymarin is a scavenger of radicals,  such as hydroxyl, superoxide, and hydrogen perox- ide (H2O2) (Ozkilic et al., 2006). This drug decreases lipid peroxidation. It  protects  liv- er cells directly by stabilizing the membrane

permeability through inhibiting lipid peroxi- dation. Silymarin alone or in the form of sup- plements with other drugs can influence the treatment process in liver diseases in many species of animals. Many medicinal, nutraceu- tical, and botanic extracts such as S-adenosyl- methionine, N-acetylcysteine, ursodeoxycho- lic acid, allopurinol, sitagliptin,  and  vitamin E have been used as cytoprotective agents in liver disease (Papich, 2016; Abo-Haded et al., 2017). Host factors such as age, sex, individu- al genetic constitution, malnutrition, especial- ly protein deficiency, disease status, and con- comitant use of other medications can affect the severity of drug-induced hepatic disease (Akbulut et al., 2015).

The prophylactic and treatment effects of silymarin administration and the role of se- rum enzyme activities in phenobarbital-in- duced hepatotoxicity in this study were rel- atively similar to the results of Avizeh et al. (2010). In their research, a single oral admin- istration of acetaminophen in cats signifi- cantly elevated serum concentrations of ALT, AST, ALP, LDH, methemoglobin, and total and direct bilirubin. In both groups that re- ceived acetaminophen plus N-acetylcysteine or silymarin, the levels of serum enzyme ac- tivities, methemoglobin, and total and direct bilirubin remained within the normal range. They stated that the major activity of both the compounds is their antioxidant property, which makes them useful in the prevention of other organ-specific toxicities related to the induction of oxidative stress (Avizeh et al., 2010). As previously stated information about the toxicity of phenobarbital is rel- atively low in cats. Most cases of drug-in- duced hepatopathy are mild and present with vague signs of lethargy and depression with or without jaundice.

In conclusion, the results of this study