How often should phenobarbital be dosed?
Phenobarbital should be given b.i.d. (q12h) in most dogs, and is rarely necessary on a t.i.d. dosing schedule (q8h). This is because the elimination half-life of phenobarbital in most dogs is much greater than 12 hours, and serum drug levels are sustained well throughout a 12-hour period. Phenobarbital does induce its metabolism over time, and induction may increase for at least three months, depending on dose. However, in our study of 33 epileptic dogs treated chronically with phenobarbital, all dogs had an elimination half-life for phenobarbital of 14 hours or longer (median 50 hours; Levitski and Trepanier, 2000). If an inadequate response (and low serum phenobarbital concentrations) is seen on b.i.d. dosing, the dog’s elimination half-life can be estimated by a clinical pharmacologist from trough and peak (three hours post dosing) serum samples. Serum half-lives less than 24 hours may warrant t.i.d. therapy to avoid clinically relevant fluctuations in serum drug levels throughout the day. In most cases, however, giving more drug b.i.d. is more likely to result in therapeutic success than changing to a t.i.d. regimen, because compliance will be more consistent on a twice-daily schedule.
Is timing important when drawing blood for serum phenobarbital concentrations?
It has long been recommended that single serum phenobarbital concentrations be measured at specific times, either just before the next dose (trough concentrations) or at three to six hours after dosing (peak concentrations). However, such timing can be inconvenient. We hypothesized that such careful timing was not actually necessary, because serum phenobarbital concentrations would be expected to fall only very slowly over a 12-hour period. To test this hypothesis, we compared trough and peak serum phenobarbital concentrations in a group of 33 epileptic dogs on maintenance phenobarbital therapy, to determine whether there were any statistically or clinically significant differences in serum drug levels throughout the day. For these dogs, mean trough, three hour, and six hour serum phenobarbital concentrations did not differ. For 30/33 dogs (91%), serum phenobarbital concentrations were in the same therapeutic category (sub-therapeutic, low-therapeutic, mid-therapeutic, or high-therapeutic) at all time points. Thus, for most dogs, careful timing of blood sampling for phenobarbital concentrations is not critical for clinical decision-making. (As an aside, remember that serum separator tubes should not be used for measurement of serum phenobarbital (Boothe et al., 1996), since the gel in these tubes can result in falsely low serum drug concentrations).
How should serum phenobarbital concentrations be interpreted?
When a subtherapeutic drug measurement is obtained, the clinical status of the patient is important in determining whether to increase the dosage (because seizures continue); maintain the dosage (because seizures are very well controlled, and no recent change in drug administration has occurred); or attempt a slow weaning off of the drug (if phenobarbital was initiated after a single seizure, and no subsequent seizures have been observed in years). Supratherapeutic concentrations (greater than 45 ug/ml) appear to be associated with a greater risk of hepatotoxicity, and it is probably wise to add a second drug (e.g., bromide) rather than increase the dose of phenobarbital further in dogs with high serum phenobarbital concentrations and poorly controlled seizures. A recent rise in serum phenobarbital concentrations without an increase in dose may signal the onset of liver insufficiency, and should prompt bile acid testing.
For dogs also treated with bromide, the therapeutic range for serum phenobarbital concentrations is lower (Trepanier et al., 1998). Nearly half of bromide-treated dogs have well-controlled seizures at phenobarbital concentrations less than 20 ug/mL. At least one out of five dogs that are started on bromide can be taken off phenobarbital, so there is no real lower end of the therapeutic range for phenobarbital in bromide-treated dogs. Phenobarbital dose reductions should instead be guided by clinical response, with the goal being to use the lowest dose of phenobarbital that continues to control seizures in combination with bromide.
Side effects of phenobarbital
Hepatotoxicity
It is well known that phenobarbital will induce increases in liver enzyme activities (SAP, ALT) in most dogs (but interestingly, not in cats). Overt hepatic failure, with jaundice, cirrhosis, and encephalopathy, is less common (Dayrell-Hart et al., 1991). The mechanism of this hepatic failure is unclear, but it is likely to be a dose- and duration-dependent toxicity with modulating factors, rather than a true drug allergy involving an antibody response. Some dogs show resolution of clinical signs with only a dose reduction, not an absolute discontinuation, of phenobarbital (Dayrell-Hart et al., 1991). It is advisable to perform bile acids every six to 12 months in dogs treated with phenobarbital, or sooner if SAP or ALT activities are greater than three times the upper range of normal, if albumin or bilirubin become abnormal, or if sedation occurs without a dosage increase in phenobarbital. In our recent study of epileptic dogs treated with bromide and phenobarbital (Trepanier et al., 1998), the initiation of bromide and subsequent reduction in the dosage of phenobarbital was associated with a reduction in abnormal liver enzyme activities and/or bile acids in all of those dogs for which liver enzyme activities were measured.
Altered thyroid testing
Phenobarbital treatment can alter thyroid testing. Phenobarbital treatment is associated with decreased total serum T4 and decreased free T4 (Kantrowitz et al., 1999; Gieger et al., 2000). This may be due to induced metabolism of T4, increased biliary excretion of T4 due to increased bile flow, or increased peripheral deiodination of T4 to T3 (seen with phenytoin). TSH concentrations may be either normal (Kantrowitz et al., 1999) or increased (Muller et al., 1998). Thus, low free T4 in a dog treated with phenobarbital can lead to a misdiagnosis of hypothyroidism. Low free T4 and total T4 have been shown to normalize by four to six weeks after discontinuation of phenobarbital (Gieger et al., 2000) at which time thyroid testing, along with serum cholesterol levels and clinical evaluation, can be repeated.
Altered adrenal testing
Phenobarbital also speeds the clearance of dexamethasone, but should not affect low dose dexamethasone suppression testing in most dogs (Chauvet et al., 1995; Mueller et al., 1998b). Phenobarbital does not affect the ACTH stimulation test, or endogenous ACTH concentrations (Dyer et al., 1994). Dogs that are treated for Cushing’s disease with mitotane (Lysodren) that are also being treated with phenobarbital may require higher loading and maintenance dosages of mitotane (Kintzer and Peterson, 1991). This is most likely due to phenobarbital-induced increases in the elimination of mitotane.
Hematologic toxicity
Blood dyscrasias associated with phenobarbital administration appear to be rare. However, phenobarbital (or primidone) has been associated in dogs with thrombocytopenia and neutropenia, with or without anemia, after one to five months of therapy (Inzana, et al., 1998; Jacobs et al., 1998). Neutropenia was severe in some dogs (100–1000 cells/ul) and was associated with fever. Thrombocytopenia ranged from < 10,000 to 149,000 plts/ul. Hematocrits ranged from 14% to 36% in anemic dogs, and were regenerative in two cases. It is difficult to determine from these reports whether the underlying pathogenesis was the same for each case, since some anemias were mild and nonregenerative, and others were severe and suggestive of hemolysis. Possibilities for the underlying mechanism include deranged folate metabolism (seen with phenobarbital in humans), drug-induced immune-mediated disease, or direct marrow toxicity due to toxic drug metabolites. A CBC is indicated in any dog treated with phenobarbital that shows signs of fever, lethargy, pallor, or bruising.
How does bromide work?
Bromide is a halide element with anticonvulsant properties that is gaining widespread use for the treatment of seizures in dogs. Bromide exerts its anticonvulsant effects by passing through neuronal chloride channels and hyperpolarizing neurons. Bromide is one of the few drugs that have been shown to be effective in dogs with seizures that are refractory to standard, therapeutic serum concentrations of phenobarbital. However, successful control of seizures in these patients is dependent upon appropriate management of bromide therapy.
When is bromide indicated?
Bromide is indicated 1) for epileptic dogs treated with either phenobarbital or primidone, that have continued seizures despite therapeutic serum phenobarbital concentrations; 2) for dogs with or without adequate seizure control, with evidence of hepatotoxicity or other undesirable side effects from barbiturates; 3) for dogs with epilepsy and evidence of pre-existing liver disease (whether or not the liver disease is related to the seizures); and 4) as a first line anticonvulsant in dogs with newly diagnosed epilepsy. In my experience, bromide as a single agent appears to be as (or more) effective than phenobarbital. Because of the risk of hepatotoxicity with phenobarbital, I prefer to start bromide as a first line agent, with smaller doses of phenobarbital added later if additional seizure control is needed.
What’s the best starting dose for potassium bromide?
The optimal starting dose for bromide in dogs being fed average commercial diets is 30–40 mg/kg once daily. Unless there is nausea associated with a single daily dose, there is no reason to give bromide twice daily. High chloride intake increases bromide loss in the urine and lowers serum bromide concentrations. Therefore, for dogs on high chloride diets (e.g., dry diets with chloride contents greater than 1.0%, such as Eukaneuba Response Formula FP Ô, and Hill's h/d, s/d, or i/d), higher starting doses of 50–80 mg/kg/day of KBr may be necessary to maintain serum bromide concentrations in the therapeutic range. For dogs on bromide alone, 40–50 mg/kg/day is a reasonable starting dose, although higher doses may be necessary to control seizures. Because bromide elimination does vary with chloride intake, all dosages should be optimized by monitoring serum bromide concentrations.
Is a loading dose of bromide necessary?
For dogs already on phenobarbital, a loading dose is probably not necessary and can cause significant sedation. A loading dose is indicated if bromide is being started alone in a dog with frequent or severe seizures, or if phenobarbital or primidone must be discontinued rapidly in the face of hepatotoxicity. I prefer that dogs be given a loading dose of bromide in the hospital so that they can be monitored and so that there is no confusion on the part of the owner about continuing the loading dose at home. A total dose of 600 mg/kg (divided into four doses of 150 mg/kg) over 24 to 48 hours achieves therapeutic serum bromide concentrations in most dogs.
How should the dose of phenobarbital be adjusted when bromide is added?
Seizure breakthrough can occur if the dose of phenobarbital or primidone is tapered too rapidly following the addition of bromide. It is safest to wait until mild sedation develops to reduce the dose of phenobarbital (usually within the first four weeks of therapy), or, if sedation does not occur, to wait until bromide concentrations are within the therapeutic range, to begin to taper the dose of phenobarbital. Dose reductions of 10–25% every four to six weeks (depending on the patient's seizure interval) work well for most dogs. Because phenytoin and oral diazepam are only minimally effective as anticonvulsants in dogs, the dosages of these drugs can be tapered more rapidly once bromide therapy has been instituted.
How should serum bromide levels be interpreted?
I recommend a therapeutic range for bromide of 1000–2500 ug/mL in my laboratory. This is equivalent to 1–2.5 mg/mL or 100–250 mg/dL. In our review of 122 epileptic dogs, dogs on both bromide and phenobarbital that had improved seizure control (defined as a 50% or greater reduction in seizure frequency following the initiation of bromide) had mean serum bromide concentrations of approximately 1600 ug/mL. Only eight out of 67 dogs that improved on combination therapy had serum bromide levels less than 1000 ug/mL. Therefore, serum bromide concentrations should be targeted to above 1000 ug/mL for most dogs on both bromide and phenobarbital.
For dogs treated with bromide alone, serum bromide concentrations need to be higher than for dogs also treated with phenobarbital. About 50% of dogs treated with bromide alone require bromide levels greater than 2000 ug/mL to show seizure improvement, while 10% of dogs require levels even greater than 3000 ug/mL. Thus, the upper end of the therapeutic range for dogs on bromide alone should be extended to approximately 3000 ug/mL, with careful monitoring for side effects.
Can most epileptic dogs on bromide be taken off phenobarbital?
At least 20% of dogs treated with bromide can tolerate a discontinuation of phenobarbital while improving seizure control. An additional 30% or more of dogs can tolerate phenobarbital dose reductions with improved seizure control. If discontinuation of phenobarbital is the goal in a particular patient, serum bromide concentrations should be targeted to greater than 2000 ug/mL.
What are the chances that a refractory epileptic patient will improve on bromide?
Most dogs with epilepsy show improvement when bromide therapy is added. Podell and Fenner reported in 1993 that 83% of their canine epileptic patients with refractory epilepsy had a reduction in seizures following the initiation of bromide, with 26% of dogs becoming seizure-free. Ninety percent of dogs treated with both phenobarbital and bromide in our study of 122 epileptic dogs (Trepanier et. al., 1998) reached serum bromide concentrations greater than 2000 ug/mL and showed significant seizure improvement. Thus, for dogs that have continued poor seizure control on bromide, serum bromide concentrations should be targeted to greater than 2000 ug/mL.
What side effects of bromide are most common and how should they be managed?
Lethargy, sedation, ataxia, and hindlimb weakness are quite common with bromide and phenobarbital in combination. These side effects are dose-dependent and usually respond within several days to a 10 to 25% reduction in the dose of phenobarbital. Although a reduction instead in the dose of bromide can also be considered, the goal in most patients is to reach therapeutic serum bromide concentrations and to reduce the dose of phenobarbital over time, so that a dose reduction in phenobarbital is preferred. If lethargy and sedation do not respond within five to seven days to a reduction in the dose of phenobarbital, serum bromide levels should be evaluated. Other side effects of bromide with or without phenobarbital include polyuria, polydipsia, polyphagia with weight gain, and behavior changes. These side effects are a nuisance but are rarely significant enough to require drug discontinuation.
REFERENCES are AVAILABLE ON REQUEST