A case of refractory secondary paroxysmal kinesigenic dyskinesia with high sensitivity to phenytoin monotherapy

1. Introduction 

The majority of cases of paroxysmal dyskinesias are idiopathic or familial in nature [1]. However, in rare cases, paroxysmal dyskinesias appear after diverse brain lesions. Anticonvulsants seem to be the most beneficial drugs for the treatment of symptomatic paroxysmal dyskinesias [2].

We describe a 42-year-old man with cerebral palsy who presented with frequent episodes of paroxysmal dystonic posturing refractory to all the antiepileptic and neuroleptic therapy that had been prescribed. The attacks took the form of paroxysmal kinesigenic dyskinesia (PKD).

2. Case history 

The patient was a 42-year-old man presenting with frequent paroxysmal hemidystonia precipitated by various stimuli. He had a history of cerebral palsy with a residual bilateral pyramidal syndrome and mild intellectual retardation. At the age of 18, during a stressful event (an earthquake) he had his first episode of a hemidystonic spasm that lasted a few seconds. Following this, he had recurrent episodes with increased frequency and duration. By the age of 24 he was unable to walk without support.

At the time of consultation he was having 10–20 episodes/day, all of the same type. They consisted of sudden dystonic posturing involving the right half of the face and right extremities. There was no alteration of consciousness. Each attack lasted up to 1min and was always triggered by a stimulus e.g., a loud noise, a stumble, or a friendly touch on his back. Until then no medication had been effective. Clonazepam, diazepam, valproic acid, phenobarbital, carbamazepine, oxycarbamazepine, aloperidol, olanzapine, anticholinergics, levetiracetam, l-dopa, and phenytoin failed to control the episodes; however, phenytoin had only been given in combination with other drugs, never as monotherapy and only up to 200mg/day.

The patient underwent a comprehensive blood and urine screen that included thyroid function tests, antithyroid antibodies, antinuclear antibodies, copper and ceruloplasmin levels that were negative or within the normal limits. MRI of the brain disclosed an old ischemic lesion between the basal ganglia and the subcortical area of the left parietal lobe. An electroencephalographic study performed at rest and during an attack did not reveal any epileptic activity.

Consistent with the history, clinical picture, imaging findings and the negative results of the laboratory screen, he was diagnosed with a diagnosis of secondary paroxysmal dyskinesia of the dystonic type. As described in the literature we decided gradually to withdraw olanzapine, levetiracetam, and valproic acid and started phenytoin per os at 100mg tid. Ten days following phenytoin monotherapy the patient had improved significantly and he was episode free for a month. In the second month he had one episode every 4–5 days. An increase of phenytoin to a total dose of 400mg/day reduced the frequency of the attacks to one episode a week. The patient was now able to walk without support. Five months later he consulted a neurologist elsewhere, who advised him to discontinue phenytoin because of potential hepatic toxicity and start gabapentin at 300mg bid. After 2 days on gabapentin use (300bid) and phenytoin (100mg tid) the patient experienced a dramatic increase in paroxysms, to 3–4 episodes/day. The gabapentin was discontinued and phenytoin was increased again to 400mg/day. This led to a sharp reduction in the frequency of the episodes to 1–2/week.

3. Discussion 

Most cases of paroxysmal dyskinesias are primary, categorized as either familial or idiopathic. Two main types of paroxysmal dyskinesias can be distinguished, PKD and paroxysmal non-kinesigenic dyskinesias (PNKD) [1]. The two types have differences concerning sex preponderance, frequency, duration, trigger stimulus of the paroxysm, precipitant event, and treatment (Table 1). However, in some cases of paroxysmal dyskinesias (the symptomatic paroxysmal dyskinesias) a specific cause can be identified [2] (Table 2). Lesions in multiple sites of the central and the peripheral nervous system have been associated with attacks of paroxysmal dyskinesias [3], [4], [5]. The pathophysiological mechanism of paroxysmal dyskinesias in these cases is unknown, but the various anatomical sites that have been linked to the development of the event makes possible that more than one mechanism can produce them [6], [7], [8]; however, the therapeutic response seems to be related to the type of the attacks (PKD or PNKD) rather than the anatomical lesion [2].

Table 1. Two main types of paroxysmal dyskinesias

		PKD	PNKD
	Manifestation age (years) range	<1–40	<1–30
	Median	12	12
	Mean	12	12
	Sex preponderance
	Males:females	4:1	4:1
	Paroxysm
	Frequency	Up to 100/day	Up to 3/d
	Duration	Up to 5/min	Up to several hours
	Trigger	Rapid voluntary movements	Nil
		Startle reaction
		Hyperventilation
	Precipitant	Stress	Fatigue
			Stress
			Excitement
			Alcohol
			Caffeine
	Treatment	Anticonvulsant	Unrewarding

Table 2. Causes of symptomatic paroxysmal dyskinesias – SPD

	Multiple sclerosis
	Myelopathy
	Head trauma
	Cerebral palsy
	Basal ganglia calcification
	Cerebral infarcts and hemorrhages
	Focal seizures
	Encephalitis
	Radiculopathy
	Hypoparathyreoidism
	Thyreotoxicosis
	Hypoglycaemia
	Reflex sympathetic dystrophy

In our case of a patient with cerebral palsy, there was an ischemic lesion between the basal ganglia and the subcortical area of the left parietal lobe. The characteristics of his attacks resemble those of PKD. They were short in duration (<5min), with a frequency of 10–20/day, triggered by rapid voluntary movements, startle reactions and emotional stress. The administration of clonazepam, valproate acid and phenobarbital, failed to control the attacks. The co-administration of a second or a third drug (phenytoin, carbamazepine, gabapentin l-dopa, neuroleptics, anticholinergic) had no effect. Some combinations even increased the frequency of the attacks as was the case when gabapentin was added to phenytoin. The administration of high doses of phenytoin (400mg/day) on monotherapy led to a dramatic reduction in the frequency and duration of the attacks.

Idiopathic and secondary cases of PKD share the same clinical features and possibly the same pathophysiological mechanism; at the present time this mechanism is unclear; however, ion channel abnormalities have been identified in other paroxysmal movement disorders giving us the opportunity to speculate that PKD is an ion channel disorder too [9]. The good response of PKD to drugs that block the voltage-sensitive sodium channels like carbamazepine, phenytoin, and lamotrigine favours that hypothesis [10]. At the same time, other antiepileptics e.g., valproate, seem to be less effective [11].

Thus, one can speculate that idiopathic PKD cases may be caused by mutations in the voltage-gated sodium ion channel genes and secondary cases may emerge from chronic injury in central and peripheral neurons. In regard to neuropathic pain, it has been shown that chronic injury to peripheral sensory neurons can produce marked modifications in the expression of genes encoding both Na+ and Ca+ channels, resulting in changes in their distribution and composition [12], [13]. Sodium channel-blocking antiepileptic agents are effective in neuropathic pain. Possibly, a similar mechanism causing alterations in the function of voltage-gated sodium channels exists in secondary cases of PKD, explaining their good response to the same antiepileptic drugs.