Refractory Seizures Associated With an Organic Aciduria in a Dog

Introduction

Levetiracetam was approved in November 1999 for use in adult humans as an add-on therapy for the treatment of partial seizures that may or may not become generalized.^1,2 Levetiracetam displays potent anticonvulsant activity in a broad range of animal models of chronic epilepsy.^3,4 Levetiracetam is not metabolized by the liver, is excreted by the kidneys, and is free of significant drug-drug interactions; therefore, it is potentially a very safe medication in dogs with seizures of any etiology.^1–4

Infrequent and newly identified causes of seizure activity in humans and dogs include inborn errors of cerebral metabolism, such as organic acidurias.^5–9 Organic acidurias represent a group of inherited disorders of deficient activity of specific enzymes that catabolize amino acids, carbohydrates, or lipids with resultant tissue accumulation of one or more carboxylic (organic) acids.^5,6 Humans affected by organic acidurias predominantly present with neurological symptoms and structural brain abnormalities—the etiopathogeneses of which are poorly understood.^5,6 More than 50 organic acid disorders have been described in people, whereas only two have been clinically recognized in dogs (i.e., L-2-hydroxyglutaric aciduria in Staffordshire bull terriers and malonic aciduria in Maltese dogs).^7–9 Variable clinical signs accompany the human disorders and often relate to cerebral dysfunction (e.g., altered mentation, seizure activity).⁵ Symptomatic and dietary therapies are the mainstays of treatment in humans, but they are not often successful.^5,10,11

The purpose of this report is to describe a new organic aciduria in a dog (i.e., hexanoylglycinuria potentially from medium-chain acyl-CoA dehydrogenase deficiency) and treatment of the associated seizures with levetiracetam.

Case Report

A 6-month-old, 10-kg, female Cavalier King Charles spaniel was evaluated by the referring veterinarian for an 8-week history of neurological signs. An intermittent head tremor had initially been noted by the owner, in conjunction with episodes of ataxia. Hypermetria of the forelimbs occurred during the ataxic episodes. Clinical signs progressed over the next 2 months, with increased frequency and severity of the episodic ataxia. The dog intermittently became stuporous and fell into lateral recumbency. While in lateral recumbency, the forelimbs twitched erratically and had increased extensor tone. Excessive blinking and facial twitching accompanied by involuntary jaw movements were also noted. Menace responses were absent bilaterally, consciousness was diminished, and no rapid eye movements occurred (as with narcolepsy). These episodes would last 5 to 10 seconds, and immediately afterward the dog appeared clinically normal. The episodes were consistent with seizure activity and occurred in excess of 50 times per day. A multifocal or diffuse intracranial lesion was considered, with the cerebrum and cerebellum as probable sites. The animal was completely vaccinated at 3 months of age and was fed a commercial puppy food, and there was no history of toxin exposure or trauma. The dog was referred for further investigation of the etiology of the seizures.

Upon presentation at 8 months of age, the dog was in good body condition (score=3/5), alert, and responsive to stimuli. General physical and neurological examination findings were unremarkable. The animal had an unusual odor that could not be localized, and it was not associated with halitosis or skin disease.

Serum biochemical profile identified hypertriglyceridemia (5.91 mmol/L, reference range <0.6 mmol/L) and mild hypoglobulinemia (27 g/L, reference range 28 to 42 g/L). Blood glucose was normal. Hematology was unremarkable. Toxoplasma gondii and Neospora caninum anti-body titers were negative, and canine distemper virus serology was consistent with prior vaccination. Urinalysis revealed the presence of some amorphous urate and phosphate material. Fasting and postprandial bile acids were normal. Serum triglyceride levels (0.98 mmol/L) repeated after the introduction of a fat-restricted diet were reduced but were not associated with decreased seizure activity.

Magnetic resonance imaging (MRI) of the brain was performed with a 1.5T Signa MR unit.^a No intracranial parenchymal abnormalities were noted on T1-weighted, T2-weighted, and contrast-enhanced images. Occipital dysplasia was found, but its significance was uncertain. Cerebrospinal fluid sampling was declined by the owner.

Potassium bromide^b (KBr, 25 mg/kg per os [PO] q 12 hours) was initiated after the MRI, and serum bromide levels were found to be within the therapeutic range (1300 mg/L, reference range 1000 to 3000 mg/L) after 3 months of therapy. Despite therapeutic serum levels, clinical response was poor. Gabapentin^c (15 mg/kg PO q 12 hours) was added at this stage. Reduced frequency and severity of the focal motor seizures were noted following initiation of gabapentin; however, gabapentin was accompanied by marked sedation and occasional loss of bladder control.

Generalized tonic-clonic seizures were reported approximately 8 months after the initial onset of clinical signs, while the dog was still on gabapentin and potassium bromide. Clusters of three to four seizures occurred every 4 to 6 weeks. The episodic ataxia appeared to have disappeared when the focal motor seizure frequency increased, although a specific time association in the change was not made.

By the time the dog was 14 months of age, each generalized seizure lasted approximately 30 to 60 seconds and was associated with extensor rigidity, involuntary muscular movements, loss of consciousness, and profuse salivation. In the post-ictal period, the dog was mentally confused and ataxic for several hours. Cluster seizures continued with increasing frequency, with the cluster events occurring closer together. Phenobarbital^d (3 mg/kg PO q 12 hours) was administered in addition to the aforementioned therapeutics. When serum phenobarbital levels were documented to be at the lower end of the therapeutic range (69 μmol/L, reference range 65 to 170 μmol/L), a reduction in the frequency of generalized seizures was noted, and cluster activity decreased to every 8 to 12 weeks. Rectal diazepam^e (0.5 mg/kg) was used in an attempt to reduce the number of seizures in each cluster. Focal motor seizures were still poorly controlled and occurred in excess of 20 times daily.

Six months after initiation of phenobarbital, it was decided to investigate the dog for underlying, intrinsic cerebral metabolic diseases based upon the signalment and the fact that the seizures were unusually refractory to therapy. Urine was evaluated for routine organic acid and amino acid levels, ^12,f and results were consistent with increased urinary excretion of hexanoylglycine [see Table]. Urine amino acid levels were considered normal.

Because of the continued poor control of the seizures, gabapentin was gradually withdrawn and replaced by levetiracetam^g (25 mg/kg PO q 12 hours). A dramatic clinical improvement occurred within 4 weeks, with a reduction in the frequency of focal seizure activity to once-weekly clusters and the generalized seizure activity to once monthly.

The dog’s seizures were well controlled on oral KBr, phenobarbital, and levetiracetam therapy for a further 28 months. At this time, the trough serum levels of levetiracetam were 1.4 mg/L (human reference range 5 to 45 mg/L). Serum phenobarbital (51 μmol/L, canine reference range 40 to 160 μmol/L) and serum bromide (1143 mg/L, canine reference range 1000 to 2500 mg/L) concentrations were within the recommended therapeutic ranges. The focal motor seizures were infrequent (i.e., every 3 to 4 weeks), and generalized seizures occurred every 3 to 4 months in clusters.

A repeat urine sample revealed the same abnormalities as previously documented [see Table]. Additionally, a plasma sample demonstrated an abnormal acylcarnitine profile, providing a strong indication of a defect in the dog’s fatty-acid oxidation mechanisms. Unfortunately, the dog suddenly exhibited several episodes of status epilepticus and was euthanized.

Discussion

The dog in this report was found to have excessive urinary levels of hexanoylglycine, which was similar to humans with medium-chain acyl-CoA dehydrogenase deficiency.¹² Medium-chain acyl-CoA dehydrogenase deficiency is the most frequently described metabolic disorder of fatty-acid oxidation in humans, and it is the main cause of what has been described as primary carnitine deficiency.¹² In this disorder, toxic levels of acyl-CoA accumulate and impair the citric acid cycle, gluconeogenesis, the urea cycle, and fatty-acid oxidation. ¹² Abnormal metabolites, such as suberylglycine, n-hexanoylglycine, 3-phenylpropionylglycine, and octanoylcarnitine are excreted in the urine, which support the diagnosis.¹²

Urinary organic acids and acylcarnitines can be quantified by gas chromatography/mass spectrometry and liquid chromatography/mass spectrometry.^13,14 The accumulating organic acids are excreted in the urine in free or esterified forms.¹³ Analysis of the abnormal excreted compounds provides a basis for the diagnosis of the genetic disorder underlying the acidotic syndrome.¹⁰ No abnormal levels of other organic acids or amino acids were detected in the urine of the dog in this report. Cultured fibroblast analysis was not performed in this dog, and no further enzymatic or genetic analysis was pursued.

One possibility considered for the biochemical abnormalities detected in this dog was interference with medium-chain acyl-CoA dehydrogenase activity by the anticonvulsants being administered. A literature search found no supporting evidence for this theory; however, valproate has been shown to inhibit the four types of acyl-CoA dehydrogenases involved in branched-chain amino acid metabolism and short- and medium-chain fatty-acid metabolism.¹⁵

It is possible that the etiology of the seizure activity in this dog was unrelated to the organic aciduria. Without a successful specific treatment for organic aciduria and a subsequent reduction in seizure frequency and/or severity, cause and effect could not be established. The dog in this report had occipital dysplasia identified on MRI, and seizure activity has been reported in association with this abnormality in both humans and dogs.^16–19 The prevalence of seizure activity with this malformation was difficult to estimate, as this breed is also predisposed to idiopathic epilepsy.¹⁶ Cavalier King Charles spaniels with idiopathic epilepsy arise from a different genetic subset than do those with occipital malformation.¹⁶ A prior study of 39 Cavalier King Charles spaniels with neurological abnormalities and occipital dysplasia documented seven (18%) dogs with seizure activity.¹⁸ The mechanism of seizure activity associated with occipital dysplasia is uncertain but may be related to cerebral microdysgenesis.¹⁹ It is possible that the seizure activity in the dog reported here was associated with occipital malformation or idiopathic epilepsy.

There are no known successful and specific treatments for organic acidurias. Dietary supplementation with L-carnitine is recommended in humans with medium-chain acyl-CoA dehydrogenase deficiency, but side effects can include nausea, vomiting, diarrhea, and abdominal cramps.^11,20 L-carnitine supplementation was not considered in this dog, and, in retrospect, it could have been administered at the time of initial diagnosis. Seizure activity associated with organic aciduria may be related to recurrent nonketotic hypoglycemia; however, no evidence of hypoglycemia was found in this dog. In the instance of recurrent hypoglycemia, symptomatic management can be achieved by avoidance of fasting. Because of the low numbers of reported cases of organic acidurias in dogs, the prevalence of seizure activity in affected animals and the success of seizure therapy are unknown.

Levetiracetam is the S-enantiomer of the ethyl analogue of piracetam, and it has broad-ranging, unique, and incompletely understood mechanisms of action against seizures.^3,4,21 Levetiracetam may decrease the onset of a seizure by counteracting a negative allosteric modulator of gamma-aminobutyric acid_A receptors, thereby causing enhanced gamma-aminobutyric acid-activated chloride ion conductance.^3,4,22 Additionally, levetiracetam has been reported to decrease calcium current production by inhibiting the N-type, high-voltage-activated calcium channels.²² The drug is well absorbed but is more rapidly metabolized than other anticonvulsant drugs in humans.¹ The pharmacodynamic therapeutic effect is believed to outlast the known half-life of the drug.²² In dogs, the drug has a half-life of approximately 4 to 6 hours, is metabolized by the liver independently of cytochrome P450, and is excreted unchanged by the kidneys.^3,4 Despite its short half-life, levetiracetam can be given three times daily.²³ Because of its expense, it was administered twice daily to the dog in this report, which may explain why seizures were not completely controlled.

Levetiracetam is the most well-tolerated antiepileptic drug in humans, with adverse reactions equal to that of placebo.^1,2,22 Overall, this drug is a highly effective adjunctive therapy in humans with seizures refractory to treatment. ^1,2,22 One study in dogs has shown the drug to be well-tolerated and effective in reducing seizure activity by up to 54% in dogs with generalized tonic-clonic seizures associated with idiopathic epilepsy.²⁴ The reason for the apparent success of this drug in the case reported here was unknown; however, the use of levetiracetam should be considered for refractory focal and generalized seizures and for seizures secondary to organic acidurias.

Conclusion

Seizures associated with an organic aciduria were diagnosed in a young Cavalier King Charles spaniel. Seizure activity was refractory to potassium bromide, phenobarbital, and gabapentin, but improved when levetiracetam was added to the first two drugs. Further investigation is necessary to establish an accurate prevalence of organic acidurias in dogs to develop specific treatments for these diseases.