Clinical

The prevalence of hypoKPP is approximately 1 per 100,000. Episodes of limb weakness accompanied by hypokalemia usually begin during adolescence. Attacks usually occur in the morning and are often triggered either by the ingestion of a carbohydrate load and high salt intake the previous night, or by rest following strenuous exercise. Generalized muscle weakness and reduced or absent tendon reflexes are characteristic. Heralding the weakness may be sensory changes, fatigue, or a feeling of heaviness or aching in the legs or back. During paralysis, level of consciousness and sensation are preserved. Paralysis either spares the facial and respiratory muscles or causes only mild weakness, making medical intervention rarely necessary. The frequency, length, and severity of attacks vary. Although attacks may occur several times a week, they more often occur at intervals of weeks or months. Attack durations vary from minutes to days, typically lasting several hours. Occasionally the attacks are sufficiently brief to cause difficulty in documenting the accompanying hypokalemia. Patients usually recover full strength, although mild weakness may persist for several days or (more rarely) be permanent. The prophylactic use of acetazolamide usually prevents the acute attack. A progressive permanent myopathy may develop later in life.

Pathophysiology

In up to 70% of cases, the responsible mutation has been linked to a gene encoding a VGCC on chromosome 1q (Venance et al., 2006). The gene, CACNA1S, encodes the α₁-subunit of the dihydropyridine-sensitive L-type VGCC found in skeletal muscle. This channel functions as the voltage sensor of the ryanodine receptor and plays an important role in excitation-contraction coupling in skeletal muscle. Four mutations in the S4 segments are responsible for voltage sensitivity. Two mutations, one involving arginine-to-histidine substitutions within the highly conserved S4 segments of DII and DIV (Arg-528-His and Arg-1239-His) account for most cases. The others involve arginine-to-glycine substitutions at the same locations.

Some 10% to 20% of families with hypoKPP have mutations in the gene encoding the α-subunit of the skeletal muscle voltage-gated sodium channel (SCN4A) on chromosome 17q. This is the same channel implicated in hyperKPP and other disorders described later (see Fig. 64.1). Evidence suggests that this sodium channel–associated syndrome is phenotypically different from the more common CACNA1S form. A proposed separate clinical entity, hypoKPP2, may be distinguishable from hypoKPP associated with CACNA1S by the presence of myalgias following paralytic attacks, and the presence of tubular aggregates instead of vacuoles in the muscle biopsy. In some patients, acetazolamide worsens symptoms (Bendahhou et al., 2001; Sternberg et al., 2001). In a large retrospective series, hypoKPP2 was associated with an older age of onset and shorter duration of attacks than classical hypoKPP (Miller et al., 2004).

Whether involving SCN4A or CACNA1S, virtually all mutations causing hypoKPP involve an S4 voltage-sensor domain. In the case of the sodium channel, these mutations allow a leak current to pass through the “gating pore” at resting membrane potentials, bypassing the central channel pore and leading to inappropriate muscle fiber depolarization and consequent channel inactivation and action potential failure (Sokolov et al., 2007). Speculation exists that this phenomenon may also occur in mutated VGCCs.

Diagnosis

An accurate medical history is essential for the diagnosis because observation of attacks is unusual, and patients are often normal between attacks. Characteristic features of hypoKPP that distinguish it from hyperKPP are that paralytic attacks are less frequent, longer lasting, precipitated by a carbohydrate load, and often begin during sleep (Table 64.2). Potassium concentrations are usually low during an attack, although concentrations less than 2 mM should suggest a secondary form of periodic paralysis. Electrocardiogram (ECG) changes such as increased PR and QT intervals, T-wave flattening, and prominent U waves suggest an underlying hypokalemia. Provocative testing can be dangerous and is not routine. Test performance requires a hospitalized setting with continuous cardiac monitoring and should be performed only in patients without cardiac or renal disease. After giving an oral glucose load (2-5 mg/kg up to a maximum of 100 g) without or, less commonly, with subcutaneous insulin (0.1 U/kg), perform serial examinations of strength while monitoring serum glucose and potassium concentrations. Other diagnostic tests include electromyography (EMG), which may show decreased compound muscle action potential amplitudes during attacks compared with interictal values. Muscle histology reveals nonspecific myopathic changes of tubular aggregates or vacuoles within fibers. Genetic testing should render muscle biopsy and provocative testing obsolete for diagnosis.

Thyrotoxic periodic paralysis may be clinically indistinguishable from hypoKPP, except that it is not familial and serum potassium levels are often lower than in familial hypoKPP (<2.5). Some cases may be associated with a mutation in KCNJ18, the gene encoding a novel inwardly rectifying potassium channel (Ryan et al., 2010). All patients with hypoKPP require screening for hyperthyroidism, as the treatment—correction of the thyroid disorder—differs from that outlined later for idiopathic hypoKPP. Exclude other secondary forms of hypokalemic paralysis when serum potassium concentrations remain low between attacks. Renal, adrenal, and gastrointestinal causes of persistent hypokalemia are common, and thiazide diuretic use or licorice (glycyrrhizic acid) intoxication are considerations.

Treatment

An effective holistic approach to treatment includes lifestyle modifications and acute and chronic pharmacological intervention. Dietary modification to avoid high carbohydrate loads and refraining from excessive exertion helps prevent attacks. Oral potassium (5-10 g load) reverses paralysis during an acute attack. Prophylactic use of acetazolamide (Table 64.3) decreases the frequency and severity of attacks. Dichlorphenamide is another carbonic anhydrase inhibitor that effectively prevents attacks, as demonstrated in a randomized clinical trial (Tawil et al., 2000) where the average dose was 100 mg daily. More potent than acetazolamide, dichlorphenamide may be useful when efficacy of the former begins to fail. Many believe, without supporting evidence, that reducing the frequency of paralytic attacks provides protection against the development of myopathy. As insight into molecular pathophysiology expands, new treatment possibilities—and a new understanding of old treatments—seem likely (Matthews and Hanna, 2010).

Table 64.3 Acetazolamide

Please note that this table is for brief informational purposes only. Prescribing physicians should consult a pharmacist or an appropriate reference for complete and updated information.

bid, Twice daily; BUN, blood urea nitrogen; CBC, complete blood cell count; qid, 4 times daily.

Clinical

Characteristic of this disorder is episodic weakness precipitated by hyperkalemia. Although the weakness is generally milder than in hypoKPP, it can be sufficiently severe in hyperKPP to cause flaccid quadriparesis. As in hypoKPP, respiratory and ocular muscles are unaffected and consciousness is preserved. Frequency of attacks varies from several per day to several per year. Attacks are usually brief, lasting 15 to 60 minutes, but may last up to days. Unlike hypoKPP, myotonia is present between attacks. Onset is usually in infancy or childhood, and characteristic attacks occur by adolescence. Triggers include rest after vigorous exercise, foods high in potassium, stress, and fatigue. Despite its name, hyperKPP is often associated with a normal serum potassium concentration during an attack (see following Diagnosis section). Most patients experience a subacute onset, and some describe paresthesias or a sensation of muscle tension prior to attacks. In these situations, mild exercise may abort or lessen the severity of the attack. Mild weakness may persist afterward, and the later development of a progressive myopathy is common.

Pathophysiology

HyperKPP is as an autosomal dominant disorder, with some sporadic cases. The disorder links to SCN4A, the same gene responsible for a minority of hypoKPP cases. Among several identified missense mutations, four account for about two-thirds of cases (see Fig. 64.1). Functional expression of naturally occurring mutations demonstrated a decrease in the voltage threshold of channel activation or abnormally prolonged channel opening or both (Bendahhou et al., 2002; Hayward et al., 1999), effectively increasing the depolarizing inward current. If sustained long enough, this would lead to inactivation of the sodium channels, transitory cellular inexcitability, and weakness.

Diagnosis

Despite advances in defining the underlying genetic mutations, a thorough medical and family history and physical examination remain the best diagnostic tools. Furthermore, genetic testing is not generally available, and the high number of causative sodium channel mutations makes widespread use of genetic testing unlikely in the near future. Serum potassium is normal between attacks and even during many attacks. Unlike hypoKPP, potassium administration may precipitate an attack. In the absence of provocative testing, the basis for diagnosis is the clinical presentation. Myotonia is present in many patients between attacks, either spontaneously or after muscle percussion, and failure to produce myotonia discriminates hypoKPP from hyperKPP (see Table 64.2). Take care not to confuse subjective muscle stiffness with objective changes. Peaked T waves on ECG suggest hyperkalemia and are an aid to diagnosis. As in hypoKPP, serum creatine kinase concentrations may be normal or elevated. Electrodiagnostic studies are useful for demonstrating subclinical myotonic discharges, not seen in hypoKPP. Nonspecific findings such as fibrillation potentials and small polyphasic motor unit potentials occur during late stages of disease.

A potassium-loading test provokes an attack but is not usually necessary and can be dangerous. Patients are placed on cardiac monitors following exercise and are given an oral potassium load of 1 mEq/kg (4-10 g KCl) in a fasting state. Weakness generally ensues 20 to 60 minutes later, and the accompanying rise in potassium occurs as a second peak 90 to 180 minutes after administration. Concomitant EMG shows decreasing amplitude of the compound muscle action potential.

Treatment

The goal of therapy is to abort the acute attacks and prevent future attacks. Acute attacks are often sufficiently brief and mild so as not to require acute intervention. In more severe attacks, aim treatment at lowering extracellular potassium levels. Mild exercise or eating a high sugar load (juice or a candy bar) may suffice, as insulin drives extracellular potassium into cells. Thiazide diuretics and inhaled β-adrenergic agonists are similarly helpful, and intravenous calcium gluconate may be useful when weakness is very severe. To prevent attacks, a diet low in potassium and high in carbohydrates may obviate the need for prophylactic drug therapy. Oral dichlorphenamide was useful for prophylaxis in one randomized controlled trial (Tawil et al., 2000). Acetazolamide (see Table 64.3) and thiazide diuretics are useful as well. Successful prophylaxis may decrease the later onset of myopathy, although direct proof of this hypothesis is lacking. Finally, myotonic symptoms are troublesome in some patients; based on the underlying pathophysiology, sodium channel blockers would seem an effective therapy, and mexiletine is commonly used for this purpose.

Clinical

The characteristic features of paramyotonia congenita (PMC) are paradoxical myotonia, cold-induced myotonia, and weakness after prolonged cold exposure. Unlike classic myotonia that shows a warm-up phenomenon (see following section on Myotonia Congenita), patients with PMC often show exacerbation of myotonia after repeated muscle contraction. Symptoms are present at birth and usually remain unchanged throughout life. Myotonia affects all skeletal muscles, although the facial muscles, especially the orbicularis oris, and muscles of the neck and hands are the most common sites of myotonia in the winter. The onset of weakness is often during the day, lasts several hours, and is exacerbated by cold, stress, and rest after exercise. Many patients are asymptomatic when warm. However, cold-induced stiffness may persist for hours even after the body warms, and percussion myotonia is present even when the patient is otherwise asymptomatic.

Pathophysiology

The cause of PMC is point mutations in the SCN4A gene on chromosome 17q, the same gene mutated in hyperKPP, potassium-aggravated myotonia, and less commonly, in hypoKPP. Mutations of the gene, which include substitutions at T1313 on the DIII to DIV linker and at R1448 on the DIV-S4 segment, cause defects in sodium channel deactivation and fast inactivation. The resting membrane potential rises from −80 up to −40 mV when intact muscle fibers cool. Mild depolarization results in repetitive discharges (myotonia), whereas greater depolarization results in sodium channel inactivation and muscle inexcitability (weakness).

Diagnosis

A family history of exercise- and cold-induced myotonia strongly supports the diagnosis of PMC. However, take care to differentiate subjective and objective changes in response to cold. When asked to close their eyes forcefully and repeatedly, affected individuals exhibit progressive difficulty with relaxation and are eventually unable to open their eyes. Serum potassium concentration may be high, low, or normal during attacks, and serum creatine kinase concentrations may be elevated 5 to 10 times normal. Electrodiagnostic studies are useful in establishing the diagnosis. EMG reveals fibrillation-like potentials and myotonic discharges that muscle percussion, needle movement, and muscle cooling accentuate. Muscle cooling elicits an initial increase in myotonia, then a progressive decrease in myotonia followed by a decrease in compound muscle action potential amplitude that correlates respectively with muscle stiffness and weakness. A reduction in isometric force of 50% or more and a prolongation of the relaxation time by several seconds after muscle cooling support the diagnosis. Muscle pathology shows only nonspecific changes, and biopsy is unnecessary.

Treatment

Symptoms are generally mild and infrequent. Direct treatment, when required, at either myotonia or weakness or both. Sodium channel blockers such as mexiletine are sometimes effective in reducing the frequency and severity of myotonia. Patients with weakness often respond to agents used to treat hyperKPP (e.g., thiazides, acetazolamide). A single case report suggests the possible use of pyridostigmine (Khadilkar et al., 2010). Cold avoidance reduces the frequency of attacks.

Clinical

Inheritance of myotonia congenita (MC) is either as an autosomal dominant (Thomsen disease) or recessive (Becker myotonia) trait. The main feature is myotonia or delayed muscle relaxation after contraction. Forceful movement abruptly initiated after several minutes of rest causes the most pronounced myotonic stiffness. The myotonia of MC displays a warm-up phenomenon in which the myotonia decreases or vanishes completely when repeating the same movement several times, in contrast to the myotonia seen in patients with PMC. The onset of Thomsen myotonia is often within the first decade, whereas the onset of Becker myotonia is generally at 10 to 14 years of age. Although myotonia can affect all skeletal muscles, it is especially prominent in the legs, where it is occasionally severe enough to impede a patient’s ability to walk or run. In rare cases, sudden noise causes sufficient generalized stiffness to make the patient fall to the ground and remain rigid for several seconds. The recessive and dominant forms share many similarities, but some clinical features help distinguish the two. In general, patients with recessive disease experience transitory bouts of weakness after periods of disuse and may develop progressive myopathy (Kornblum et al., 2010); in addition, muscle hypertrophy and disease severity are greater than in the dominant form. Becker myotonia is more common than Thomsen disease.

Pathophysiology

Electrical instability of the sarcolemma leads to muscle stiffness by causing repetitive electric discharges of affected muscle fibers. Early in vivo studies in myotonic goats revealed greatly diminished sarcolemmal chloride conductance in affected muscle fibers. This causes a depolarization of the sarcolemmal membrane and muscle hyperexcitability. Genetic linkage analysis for both recessive and dominant forms of MC pointed to a locus on chromosome 7q, where the responsible gene, CLCN1, encodes the major skeletal muscle chloride channel. More than 70 mutations have been identified within CLCN1, and interestingly, some of these mutations are recognized to cause both dominant and recessive forms (Zhang et al., 2000). Examination of the functional effects of several myotonia-causing CLCN1 mutations in heterologous expression systems reveal effects on channel gating, usually resulting in a decreased chloride conductance (Zhang et al., 2000).

Diagnosis

Tapping the belly of a muscle with a percussion hammer can elicit myotonia that lasts for several seconds. Myotonia is a nonspecific sign found in several other diseases including myotonic dystrophy 1, myotonic dystrophy 2 (proximal myotonic myopathy), PMC, and hyperKPP (see earlier section). Cardiac abnormalities, cataracts, skeletal deformities, and glucose intolerance are not components of MC, and their presence should suggest one of the dystrophic myotonias. Muscle strength and tendon reflexes are normal, but patients commonly display muscle hypertrophy, often giving these patients an athletic appearance. The finding of decremental compound muscle action potential amplitudes with muscle cooling on EMG distinguishes PMC from MC. EMG in MC typically reveals bursts of repetitive action potentials with amplitude (10 µV to 1 mV) and frequency (50-150 Hz) modulation, so-called dive-bombers, in the EMG loudspeaker. Biopsy is usually nonspecific, showing enlarged fibers in hypertrophied muscle, increased numbers of internalized nuclei, and decreased type 2B fibers.

Treatment

Many patients experience only mild symptoms and do not require treatment. For those with more severe myotonia, sodium channel blocking agents remain the mainstay of treatment. Mexiletine is the most commonly used and was shown effective in treating myotonia in a randomized placebo-controlled study. Other sodium channel blockers such as tocainide, phenytoin, procainamide, and quinine exhibit variable degrees of efficacy.

Clinical

Potassium-aggravated myotonia (PAM) is a rare autosomal dominant disorder with clinical features similar to MC, except that the myotonia fluctuates and worsens with potassium administration. Distinguishing PAM from other nondystrophic myopathies is important because PAM patients respond to carbonic anhydrase inhibitors. Episodic weakness and progressive myopathy do not occur. Symptom severity varies, with some patients experiencing only mild fluctuating stiffness, and others a more protracted painful myotonia. PAM now encompasses the conditions previously known as myotonia fluctuans, myotonia permanens, and acetazolamide-sensitive myotonia. Exercise or rest after exercise, potassium loads, and depolarizing neuromuscular blocking agents aggravate myotonia, whereas cold exposure has no effect. Prominent myotonia of the orbicularis oculi and painful myotonia suggest the diagnosis.

Pathophysiology

PAM links to chromosome 17q, where mutations in the SCN4A gene cause the disease. This sodium channel is the same one implicated in hyperKPP, PMC, and the sodium channel subtype of hypoKPP (see earlier discussion and Fig. 64.1). Functional expression studies reveal that the disease-causing mutations lead to a large persistent sodium current secondary to an increased rate of recovery from inactivation and an increased frequency of late channel openings (Wu et al., 2001). The cause of myotonia is this enhanced inward current, which leads to prolonged depolarization and subsequent membrane hyperexcitability.

Diagnosis

Diagnosis is clinical because screening for the mutated gene is not widely available. Distinguishing PAM from other episodic nondystrophic myotonias may be difficult. Unlike hyperKPP and PMC, PAM patients do not experience weakness. Another distinction between PAM and PMC is the lack of response to muscle cooling, either clinically or on EMG. Furthermore, the myotonia with PAM improves with carbonic anhydrase inhibitors, whereas mexiletine is more effective in alleviating the myotonia in MC and PMC.

Treatment

Carbonic anhydrase inhibitors markedly reduce the severity and frequency of attacks of myotonia. Acetazolamide is most commonly used (see Table 64.3).

Clinical

Andersen-Tawil syndrome (ATS) is a rare autosomal dominant disorder characterized by the triad of periodic paralysis, cardiac arrhythmias, and dysmorphic features (including hypertelorism, micrognathia, low-set ears, high-arched or cleft palate, short stature, and clinodactyly) (Yoon et al., 2006a). The periodic paralysis, often triggered by rest after exercise, prolonged rest, and stress, is often the presenting symptom and can be hypo-, normo-, or hyperkalemic. The cardiac phenotype, often discovered later, includes prolonged QT intervals, but bidirectional ventricular tachycardia is common. Despite the known association of cardiac arrhythmias in rare periodic paralysis patients, ATS was only recognized as a separate entity in 1971. In families segregating an ATS allele, the phenotypic expressivity can vary greatly. Patients can manifest one, two, or three features of the triad, and the severity of any one feature can be extremely variable. Rare individuals are asymptomatic disease-gene carriers. Finally, ATS patients may exhibit neurocognitive deficits in executive function and abstract reasoning (Yoon et al., 2006b).

Pathophysiology

Mutations in the KCNJ2 gene on chromosome 17q account for approximately two-thirds of ATS probands. KCNJ2 encodes a widely expressed inwardly rectifying potassium channel (Plaster et al., 2001). Interestingly, among all identified probands, about 50% have an autosomal dominant disorder, and identification of sporadic cases with de novo mutations is common. The mechanisms of channel dysfunction are heterogeneous, including impaired phospholipid binding, pore function, or protein trafficking. Because VGKCs are tetrameric complexes, many (if not all) of the mutations are dominant negative.

Diagnosis

Previous studies that took into account the variable penetrance of ATS classified individuals as affected if two of three criteria were met: paroxysmal weakness, prolonged QT interval with or without ventricular dysrhythmias, or characteristic dysmorphic features (Yoon et al., 2006a). ATS should be included in the differential diagnosis of any individual with documented long-QT syndrome, even in the absence of periodic paralysis or dysmorphism. Some family members of patients with the full clinical triad show only prolonged QT intervals. Similarly, perform ECG on all patients with suspected periodic paralysis for careful measurement of the QT interval.

Often the diagnosis of ATS is established when an individual with paroxysmal weakness presents during an acute phase with ECG abnormalities. However, one often overlooks the diagnosis because the cardiac abnormalities, as with periodic muscle weakness, can be transitory and missed on routine ECG. A Holter monitor can capture episodic dysrhythmias or longer tracings for QT-interval analysis when ATS is suspected and the standard ECG is normal. Cardiac monitoring under provocative maneuvers such as tilt-table testing or graded exercise protocols may allow dysrhythmias to surface and should be utilized when the diagnosis of ATS is highly suggested but prior cardiac evaluation is unremarkable.

Creatine kinase concentrations are often normal during episodes of weakness but are occasionally elevated. Depending on the particular mutation, potassium concentrations during an attack are usually low but can be high or normal. Provocative testing in any case of periodic paralysis may be risky, and hypokalemic challenges in ATS may be particularly dangerous because of the potential to exacerbate preexisting QT prolongation and trigger life-threatening ventricular arrhythmias.

Treatment

The goal of therapy is to control underlying cardiac arrhythmias and to decrease the frequency, severity, and duration of attacks of weakness. Unfortunately, some treatments for arrhythmias can produce weakness, and treatments for weakness may exacerbate cardiac dysrhythmias. The periodic paralysis usually responds to carbonic anhydrase inhibitors. In ATS with hypokalemic weakness, oral potassium ingestion treats acute attacks. Prophylaxis with daily sustained-release potassium tablets may prevent attacks of weakness. Maintaining a high serum potassium level (>4.5 mEq/L) also has the additional advantage of narrowing the QT interval, thus reducing the likelihood of developing ventricular arrhythmias.

Treatment of the prolonged QT interval depends on the severity of the underlying arrhythmias. The use of beta-blockers is a mainstay of treatment in long QT. Clinical experience shows that patients with ATS tolerate these agents well. In the presence of syncope due to sustained ventricular tachycardia, the placement of an implantable defibrillator is useful. Some evidence suggests that flecainide may be effective in the treatment of severe ATS-associated ventricular arrhythmias (Bökenkamp et al., 2007; Pellizzón et al., 2008).

Clinical

Familial hemiplegic migraine (FHM) is a rare autosomal dominant subtype of migraine characterized by lateralized motor weakness of variable intensity—hemiparesis to hemiplegia. Three subtypes, genetically defined (see later discussion), are distinguishable by clinical characteristics. FHM type 1 (FHM1) accounts for about 50% to 75% of families. In addition to weakness, auras always involve additional symptoms including sensory, visual, and language disturbances (Ducros et al., 2001). Severe aura attacks may last days or weeks and may involve fever, meningismus, and impaired consciousness ranging from confusion to coma. Recovery between attacks is typically complete, although 30% to 50% of patients with FHM1 exhibit permanent progressive cerebellar signs that include nystagmus, gait or limb ataxia, or dysarthria. These manifestations of FHM1 overlap with episodic ataxia type 2 (EA2) and spinocerebellar ataxia type 6 (Fig. 64.5), all caused by mutations in the same gene (see following discussion). By contrast, the characteristic features of FHM2 are an absence of cerebellar signs and a tendency for lower penetrance than in FHM1. Ataxia does not occur in FHM3, a rarer and more recently defined entity. Importantly, patients with FHM exhibit a spectrum of disease expression, and subtype classification is likely to evolve with our knowledge of underlying genetics.

Fig. 64.5 Venn diagram showing phenotypic overlap among various channelopathies. AHC, Alternating hemiplegia of childhood; EA, episodic ataxia; FHM, familial hemiplegic migraine; SMEI, severe myoclonic epilepsy of infancy.

Pathophysiology

Familial hemiplegic migraine is a genetically heterogeneous condition that links to three loci on chromosomes 1q, 2q, and 19p; other loci are possible (Ducros et al., 1997). FHM1 includes the 50% to 75% of families that show genetic linkage to chromosome 19p13, where missense mutations in the CACNA1A gene cause the disease (Ophoff et al., 1996). This gene encodes the pore-forming α_1A-subunit of the neuronal P/Q-type VGCC, which is distributed widely throughout the brain. It is present at motor nerve terminals and the neuromuscular junction and is the principal calcium channel expressed by cerebellar Purkinje and granular neurons. These channels play an integral role in the action potential–triggered presynaptic calcium influx at nerve terminals that triggers vesicular fusion. At least 17 different missense mutations and one nonsense mutation (Jen et al., 1999) are reported in CACNA1A (Fig. 64.6). These cause a variety of altered but not lost functions: changes in channel conductance, kinetics of inactivation, and current density (Hans et al., 1999; Kraus et al., 2000). Indeed, these mutations may confer gain of function, because many appear to cause a hyperpolarizing shift in the activation voltage, meaning that channels open and permit calcium influx at abnormally low membrane potentials. In one extreme case, the S218L mutation confers both the tendency to open near the resting potential of many neurons (−60 to −50 mV) and overall slowing and reduction in channel inactivation, thus causing marked increases in overall calcium influx. In keeping with this extreme biophysical profile, S218L produces a severe clinical phenotype of aura attacks triggered by minor head trauma and leading to deep coma and prolonged cerebral edema (Tottene et al., 2005). On the other hand, some loss-of-function mutations associated with episodic ataxia type 2 (EA2) may also cause migraine symptoms (Jen et al., 2004), suggesting that a direct correlation between presynaptic VGCC-mediated calcium entry and disease severity is too simplistic. Importantly, the mechanism linking altered channel biophysics to disease expression is controversial and defies a simple model.

Fig. 64.6 P/Q-type calcium channel and the mutations causing familial hemiplegic migraine type 2. Voltage-gated calcium channels are classified into transient (T-type), long-lasting (L-type), N (neuronal), P/Q (Purkinje cell), and R (toxin-resistant) channels. Shown is the CACNA1A-encoded α_1A-subunit, which forms the voltage sensor and pore of the channel. Disease-causing missense mutations are illustrated.

As mentioned earlier, other mutations in CACNA1A cause different dominantly inherited neurological disorders, including EA2 and spinocerebellar ataxia type 6 (SCA6). Although most patients with FHM1 also suffer persistent cerebellar deficits, a few mutations in CACNA1A cause only hemiplegic migraine (Ducros et al., 2001), and one mutation appears to cause classical migraine with aura but no weakness (Serra et al., 2010). These observations suggest that a spectrum of disease exists from pure ataxia at one extreme to classical migraine at the other.

Patients with FHM2, representing 10% to 20% of all FHM sufferers, exhibit mutations in the ATP1A2 gene encoding the α₂-subunit of the Na⁺/K⁺-ATPase, a protein responsible for maintaining the membrane potential in neurons and other cells (De Fusco et al., 2003). This pump plays a critical role in establishing transmembrane ionic gradients and is in this way directly integral to the function of innumerable ion channels and other proteins. Many missense mutations have been identified, probably resulting in a loss of function but not loss of surface expression (De Fusco et al., 2003), possibly by a reduced affinity for potassium (Segall et al., 2004). Although an ionic mechanism is likely, the precise pathophysiological connection to hemiplegic aura or head pain remains obscure. Additional mutations in ATP1A2 may also cause a form of benign familial infantile seizures (see following section on Epilepsy) (Vanmolkot et al., 2003) and alternating hemiplegia of childhood (Swoboda et al., 2004). A single mutation was found to cause hemiplegic migraine with cerebellar findings (Spadaro et al., 2004), challenging the conception this association is specific for CACNA1A mutations.

FHM3 involves mutations in the SCN1A gene on chromosome 2q24 that encodes the neuronal voltage-gated sodium channel α₁-subunit. To date, three missense mutations have been identified, and the two that have been characterized electrophysiologically each confer upon the channel more rapid recovery from fast inactivation and thus the potential for faster firing frequency and neuronal hyperexcitability (Castro et al., 2009; Dichgans et al, 2005). Other mutations in SCN1A cause some cases of severe myoclonic epilepsy of infancy and generalized epilepsy with febrile seizures plus (see following section on Epilepsy). Interestingly, mutations causing epilepsy occur nearby those causing FHM3, and one mutation—L263V—causes both clinical phenotypes (Castro et al., 2009). Epilepsy has also been reported in cases of FHM1 and FHM2, pointing to the intriguing possibility of a pathogenic link between seizure and migraine.

Diagnosis

Genetic testing is not widely available, so the diagnosis relies on a thorough history and physical examination. Family history is clearly helpful in demonstrating autosomal dominance, but it is important to note that typical migraine syndromes may also show a strong (if less regular) familial pattern. Certain clinical elements of FHM help distinguish it from classical migraine with aura. Although the aura symptoms may be similar to those in classical migraine, FHM patients are more likely to experience motor, speech, and sensory symptoms. Furthermore, the duration of the headache, as well as the duration of the visual and sensory components of the aura, are typically greater in FHM patients (Thomsen et al., 2002). Although patients with classical migraine often experience the aura in isolation from the headache, FHM patients experience an associated headache virtually all the time. Infrequently, FHM patients may experience bilateral weakness, whereas this is generally not the case in patients with classical migraine. Finally, although defined genetically, one might clinically suspect FHM based on an association with cerebellar abnormalities or seizures (see Fig. 64.5).

Treatment

Anecdotal evidence suggests that acetazolamide reduces the frequency of migraine attacks (Battistini et al., 1999; Jen et al., 2004). As suggested by functionally enhanced calcium currents in many cases of FHM, the calcium channel blocker, verapamil, aborted an attack when administered intravenously (Yu and Horowitz, 2001). To date, no controlled study has tested the efficacy of these or other agents in FHM. The guidelines for treating common migraine may be applied, except that triptans and ergotamine are not used, out of concern—possibly unwarranted—for stroke (Artto et al., 2007). Nimodipine is contraindicated because of the risk of worsening symptoms (Mjåset and Russell, 2008).

Clinical

Characteristic of EA1 are attacks of cerebellar incoordination with jerking limb movements, often accompanied by slurred speech, that last for seconds to minutes. The episodes can occur spontaneously, but common triggers include exertion, infection, stress, or startle. Between attacks, patients may show myokymia, muscle rippling resulting from motor nerve hyperexcitability, especially in the hands and around the eyes. Symptom onset is in infancy, with spontaneous resolution in the second to third decade. Frequency, duration, and intensity of attacks vary greatly.

Patients with EA2 experience episodes of truncal ataxia lasting hours to days, precipitated by exertion and stress. Age of onset varies between childhood and young adulthood but is most frequently in the second decade. Vertigo, nausea, and vomiting are present in more than half of patients, and many exhibit spontaneous nystagmus that is not seen interictally. Between episodes, the patient returns to normal but frequently displays gaze-evoked nystagmus with features typical of rebound nystagmus (Jen et al., 2007). Less commonly, positional and, later in the disease course, spontaneous downbeat nystagmus occurs. Approximately half of EA2 patients report headaches that meet criteria for migraine. Since mutations causing EA2 are the same gene as in FHM1, this is expected.

Pathophysiology

The mutation underlying EA1 is on chromosome 12. At least 15 missense mutations have been described in the responsible gene, KCNA1, which encodes a delayed rectifier VGKC that opens with a delay after membrane depolarization, allowing K⁺ efflux and membrane repolarization (see Fig. 64.2). Coexpression of mutant and wild-type channels results in delayed outward potassium current and impaired membrane repolarization following an action potential (Zerr et al., 1998), a dominant-negative effect that could lead to increased neuronal excitability and neurotransmitter release. The delayed rectifier potassium channel is widely expressed in the nervous system, with highest levels in the cerebellum and myelinated axons of peripheral nerves. In the cerebellum, an imbalance between inhibition and excitation could result in brief episodic incoordination. Similarly, in the peripheral motor nerves, impaired repolarization could lead to repetitive neuronal activity and resultant myokymia. Suggesting that these two effects may be mechanistically distinct is a KCNA1 missense mutation that causes an episodic ataxia more akin to EA2 without myokymia (Lee et al., 2004a), and another missense mutation resulting in myokymia without ataxia (Rea et al., 2002).

Mutations in CACNA1A cause EA2 (Ophoff et al., 1996). Whereas FHM1 is associated with missense mutations in the same gene, the genetic alterations associated with EA2 seem more dramatic. Of the approximately 30 mutations described so far, most are truncating, a few missense, one insertional, and one due to expansion of a triplet CAG domain. Functional expression studies of some of these mutations reveal loss of voltage sensitivity or complete loss of channel function (Guida et al., 2001), suggesting that haploinsufficiency may underlie the phenotype. Some mutations may additionally interfere with protein folding and trafficking (Wan et al., 2005).

EA5 results from mutation of the VGCC β₄-subunit gene CACNB4, also implicated in idiopathic generalized epilepsy (see following section on Epilepsy); it is clinically similar to EA2 except that seizures are an additional feature (Escayg et al., 2000a). Mutations in SLC1A3, the gene encoding the excitatory amino acid transporter EAAT1, cause EA6, an episodic ataxia associated with hemiplegia and seizures (Jen et al., 2005). The genes for EA3 (1q42), EA4, or EA7 (19q13) have not been identified.

Diagnosis

Family history is helpful in making the diagnosis, despite rare cases of de novo mutations. The probability of examining a patient during an attack is low, so careful examination for interictal signs are important. Although almost half of EA1 patients complain of diplopia, they do not have nystagmus, which helps distinguish them from EA2 patients. The brevity of attacks and the persistent interictal myokymia seen in EA1 also help distinguish this from EA2. EA2 patients may demonstrate interictal end-point tremor or impaired suppression of the vestibulo-ocular reflex. They tend to have subtle and slowly progressive interictal cerebellar signs, particularly gaze-evoked nystagmus. Although progressive ataxia often develops, it is rarely severe enough to prevent walking without assistance. Magnetic resonance imaging (MRI) may detect cerebellar atrophy, especially of the anterior vermis (Fig. 64.7), whereas EA1 patients have no cerebellar atrophy. Interestingly, there appears to be an increased incidence of epilepsy in both EA1 and EA2 patients, emphasizing the overlapping symptoms possibly due to shared mechanisms (see Fig. 64.5).

Fig. 64.7 Sagittal magnetic resonance image from a patient with episodic ataxia type 2. Note the significant cerebellar atrophy from this relatively advanced case. Type 1 patients do not display cerebellar atrophy (not shown).

Treatment

Acetazolamide reduces the severity of attacks. Response to treatment varies among families, but generally, patients with EA2 have a greater response than those with EA1. Carbamazepine or 4-aminopyridine (Strupp et al., 2004) are beneficial in some patients. Because stress and strenuous exercise often exacerbate attacks, lifestyle modification can be quite effective.

Clinical

Human startle disease, or hereditary hyperekplexia, is a rare hereditary disease characterized by an exaggerated startle reflex. The usual inheritance is autosomal dominant, but several recessive mutations exist. The normal startle response is a primitive reflex that manifests as a stereotyped sequence of blinking, grimacing, neck flexion, and arm abduction and flexion. Patients exhibit an overreaction to unexpected visual, auditory, or tactile stimuli with sudden generalized myoclonic jerks followed by stiffness, often resulting in uncontrolled falling during standing and walking. Consequently, patients often develop a characteristic slow, wide-based, cautious gait. Consciousness is preserved during the attacks, which helps distinguish this from startle epilepsy. Attack frequency may increase during times of stress, fear, lack of sleep, or the expectation of being frightened. The onset of symptoms may be as early as the neonatal period, with rigidity or generalized hypertonia, nocturnal limb jerking, and an exaggerated startle response. Attacks vary in severity and frequency and may be so severe as to cause apneic episodes and even death. Affected children may show slight delay in motor development. A minor form of hyperekplexia, less common than the major form, manifests as an exaggerated startle response without associated symptoms such as neonatal stiffness.

Pathophysiology

Pathogenic mutations involve the GLRA1 gene on chromosome 5q encoding the glycine receptor α₁-subunit. The glycine receptor is a heteropentameric ligand-gated ion channel composed of three ligand-binding α₁-subunits and two β-subunits, mediating fast inhibitory neurotransmission in the brainstem and spinal cord. Several dominant, recessive, and de novo GLRA1 missense point mutations account for familial hyperekplexia patients. Most mutations flank the M2 pore-forming domain (Fig. 64.8). The physiological consequence of mutations in the α₁-subunit is decreased glycine sensitivity, impaired channel opening, and uncoupling of agonist binding from channel activation. These changes reduce glycinergic inhibition and increase neuronal excitability.

Fig. 64.8 The α₁-subunit of the human glycine receptor and pathogenic mutations underlying hyperekplexia. The glycine receptor shares significant homology with the nAChR, possessing five subunits to form a channel pore.

Mutations in the gene encoding the glycine receptor β-subunit (GLRB) also cause hyperekplexia. An example is a compound heterozygous patient with spontaneous hyperekplexia possessing a missense mutation on one allele and a splice-site mutation on the other (Rees et al., 2002). Electrophysiological studies showed reduced sensitivity of the mutant channel to agonist, suggesting impaired binding of glycine to the channel.

Finally, loss-of-function mutations in GLYT2 (also named SLC6A5), the gene encoding the neuronal glycine transporter 2 protein, have been identified in three families with hereditary hyperekplexia. Glycine transporters mediate synaptic reuptake of the neurotransmitter, and its genetic deletion reproduces hyperekplexia in mice (Eulenburg et al., 2006). Reduction of glycine transporter function is expected to prolong glycine neurotransmission; how this leads to hyperekplexia is unknown.

Diagnosis

Physical examination may reveal diffuse hyperreflexia. The diffuse hypertonicity in infancy generally resolves with time, and adults have normal tone between attacks. An exaggerated head-retraction reflex is common in these patients. Tapping the forehead or root of the nose downward with a reflex hammer causes a brisk involuntary backward jerk of the head. This reflex is generally absent in unaffected individuals. Distinguish this condition from startle epilepsy, a rare seizure disorder characterized by startle-induced tonic spasm of a limb followed by a complex partial seizure; asymmetrical tonic posturing occurs during spells, and patients often have developmental delay and focal neurological signs. Neuroimaging in startle epilepsy reveals cortical dysplasia, although an interictal electroencephalogram (EEG) rarely shows a clear seizure focus. By contrast, familial hyperekplexia patients typically have normal development, display no ictal EEG findings to suggest a seizure disorder, have normal brain imaging, and maintain full consciousness during attacks. Brainstem pathology, including pontine hemorrhage or infarction, multiple sclerosis, vascular brainstem compression, and brainstem encephalitis, may cause a syndrome similar to hyperekplexia. Rarely, sporadic cases of hyperekplexia occur; therefore, do not exclude patients lacking a family history.

Treatment

Treatment with benzodiazepines helps reduce neonatal hypertonia and significantly reduces the severity and frequency of startle-induced attacks in some patients. Although clonazepam is the standard treatment, low-dose clobazam is effective in the treatment of hyperekplexia and well tolerated in infants. Benzodiazepines act by increasing GABA-mediated inhibition and have no effect on glycinergic transmission. This suggests that enhancing the GABA-mediated inhibition may compensate for glycinergic dysfunction.

Autosomal Dominant Nocturnal Frontal Lobe Epilepsy

A syndrome characterized by clusters of brief partial seizures that occur during light sleep, autosomal dominant nocturnal frontal lobe epilepsy (ADNFLE) is a monogenic disorder with a penetrance rate of 70% to 80%. The motor seizures, which manifest as hyperkinetic tonic stiffening and clonic jerking movements, may occur several times per night, usually occurring shortly after falling asleep or just before awakening. Aura may precede seizures, manifesting as various somatosensory, sensory, and psychic phenomena. Episodes often start with a gasp, grunt, or vocalization followed by eye opening or staring. Secondary generalization is unusual, and patients remain conscious during the seizures. Patients become symptomatic within the first or second decade of life, although later onset occurs. Seizures generally persist throughout adult life, becoming less severe beyond the fifth decade. Suggesting an important contribution of genetic background, intrafamilial variability in seizure frequency and severity is significant, with some patients experiencing several seizures nightly, and others remaining seizure-free for months. Interictal EEG is normal, and ictal EEG may show bifrontal epileptiform discharges. Because of the clinical similarities, ADNFLE is often mistaken for benign nocturnal parasomnia or night terror. Therefore, nocturnal video polysomnography is very useful in distinguishing ADNFLE from these other conditions.

ADNFLE was initially shown to be caused by a point mutation in the nAChR α₄-subunit gene CHRNA4 on chromosome 20q (Steinlein et al., 1995). Four mutations in CHRNA4 were identified in several ethnic groups. Mutations have also been identified in CHRNB2 (Phillips et al., 2001) and CHRNA2 (Aridon et al., 2006), encoding the nAChR β₂– and α₂-subunits. The β₂-subunit associates closely with the α₄-subunit, and the α₄– β₂ combination is the dominant subtype of nAChR in the brain. Mutations in CHRNA4 and CHRNB2 involve M2, the second transmembrane domain and the part of the protein thought to line the channel pore. Some cases of ADNFLE link to chromosome 15q24, close to a cluster of three other nAChR subunits, although the responsible mutations have not been elucidated. Mutations result in increased sensitivity both to activation by acetylcholine and to block by carbamazepine. This latter in vitro observation bears clinical relevance because ADNFLE patients have a good therapeutic response to carbamazepine. Given the wide distribution of nAChRs within the brain, it remains unclear why the epilepsy is focal and why the seizures arise selectively in the frontal lobes.

Familial Temporal Lobe Epilepsies

Temporal lobe epilepsies exist in both sporadic and inherited forms, and the genetics of familial syndromes remains largely unknown (Andermann et al., 2005). The onset of familial lateral temporal lobe epilepsy (FLTLE), or autosomal dominant partial epilepsy with auditory features (ADPEAF), a benign syndrome distinguished from the more common mesial form by characteristic auditory auras, is in the second or third decade, and transmission is autosomal dominant with incomplete penetrance. Approximately 50% of cases involve mutations in the leucine-rich, glioma-inactivated gene 1 (LGI1) on chromosome 10q (Kalachikov et al., 2002; Morante-Redolat et al., 2002), whose product was shown to complex with presynaptic A-type VGKCs (Schulte et al., 2006). Normally, LGI1 appears to reduce channel inactivation, and its dysfunction results in faster inactivation kinetics and, likely, neuronal hyperexcitability.

Benign Familial Neonatal Seizures

Benign familial neonatal seizures (BFNS) is a rare autosomal dominant disorder. Multifocal or generalized tonic-clonic convulsions appear after the third day of life. Myoclonic seizures are rare. Seizures are generally brief and well controlled by antiepileptic medications, although status epilepticus occurs. Age of onset may extend up to the fourth month of life, and in most cases, seizures disappear spontaneously after a few weeks or months. Although these children usually have normal neurological examination and development, the risk of recurring seizures later in life is about 15%. These later seizures, often provoked by auditory stimuli or emotional stress, are easily controllable with antiepileptic medications. Interictal EEG activity is usually normal, whereas the ictal EEG attenuates at the onset, followed by slow waves, spikes, and a burst-suppression pattern.

Two BFNS-causing genes have been identified, one on chromosome 20q and the other on chromosome 8q. The mutated genes both encode VGKCs: KCNQ2 and KCNQ3, respectively (Biervert et al., 1998; Charlier et al., 1998; Singh et al., 1998). These channels activate by membrane depolarization and contribute to the repolarization of the action potential. Reports appear of missense amino acid deletion, splice-site frameshift mutations, and gene deletions; most involve the KCNQ2 gene (Fig. 64.9). Functional expression of mutant channels results in reduced potassium current, likely leading to impaired membrane repolarization and thus increased neuronal excitation. KCNQ2- and KCNQ3-encoded products combine to form a heteromeric channel underlying the M-current (Wang et al., 1998), a potassium conductance found widely in the central nervous system that plays a crucial role in the regulation of neuronal excitability. Coexpressing the mutant gene (KCNQ2 or KCNQ3) together with its wild-type allele and its wild-type partner results in mild (25%) reduction in M-current amplitude. Although it is unclear how the mutated channels either independently or in the form of the M-current lead to seizures, KCNQ2 channels are concentrated in the septum and hippocampus, areas key for control of rhythmic brain activity and neuronal synchronization, and both associated with the generation of epileptic seizures. Thus, even slight alterations in neuronal excitability through impaired KCNQ2/KCNQ3 channel-mediated repolarization could presumably lead to aberrant neuronal synchronization and seizures. In light of reports that lamotrigine and carbamazepine enhance neocortical potassium currents in vitro, they may be of particular use in BFNS patients.

Fig. 64.9 The neuronal potassium channel α-subunit encoded by the KCNQ2 gene, and the proposed pathogenic mutations causing benign familial neonatal seizures. The majority of mutations have been identified in this channel, which is believed to coassemble with the KCNQ3 gene product and underlie the M-current.

Generalized Epilepsy with Febrile Seizures Plus

Febrile seizures are the most common seizure disorder in children, affecting 2% to 5% of all children younger than 6 years. Although most febrile seizures show complex inheritance, a small proportion transmits in an autosomal dominant pattern. The disorder termed generalized epilepsy with febrile seizures plus (GEFS+) refers to the phenotype of individuals who have febrile seizures extending beyond 6 years of age, with or without afebrile generalized tonic-clonic seizures. Although most patients experience only febrile or febrile seizures plus, approximately 30% of patients may experience other generalized epilepsy phenotypes such as absence, myoclonic, and atonic spells, and even partial seizures with secondary generalization. More severe phenotypes include myoclonic-astatic epilepsy and severe myoclonic epilepsy of infancy. EEG may show irregular 2.5- to 4-Hz generalized spike-and-wave or polyspike-wave discharges.

A high level of genetic heterogeneity in GEFS+ exists (see Fig. 64.1). Mutations in four voltage-gated sodium channel genes (SCN1B on chromosome 19q and SCN1A, SCN2A, and SCN9A on chromosome 2q) (Escayg et al., 2000b; Singh et al., 2009; Sugawara et al., 2001; Wallace et al., 1998) and two GABA_A receptor genes (GABRG2 and GABRD) encoding the γ₂– and δ-subunits (Baulac et al., 2001; Dibbens et al., 2004) cause GEFS+ (Fig. 64.10). Evidence exists for other mutated genes not yet identified. Functional analysis of several mutated sodium channels reveals slow inactivation, enhanced inward sodium current, and thus neuronal hyperexcitability (Lossin et al., 2002). Functional studies of the two GABRG2 mutations showed reduced channel conductance in one case and abolished benzodiazepine sensitivity in the other. The GABRD mutations reduce current amplitude, each leading in different ways to a decrease in synaptic inhibition and thus an increase in neuronal excitability. These functional observations evoke compelling molecular explanations for clinical disease and, like ADNFLE and BFNS, illustrate how heterogeneous genetic defects may converge on a single clinical phenotype.

Fig. 64.10 Mutations in the voltage-gated Na⁺ channel causing generalized epilepsy with febrile seizures plus (GEFS+). The α-subunit is encoded by SCN1A (A), and the β₁-subunit (B) is encoded by the SCN1B gene. Disease-causing mutations are shown. Mutations in other genes, including SCN2A and GABRG2, also cause GEFS+ (see Fig. 64.1).

SCN2A Mutation and Other Early Childhood Seizures

Other rare convulsive disorders of early childhood include benign familial infantile seizures and benign familial neonatal-infantile seizures, differentiated from BFNS by the age of onset. At least some cases involve mutations in the sodium channel α₂-subunit gene, SCN2A, the same gene affected in some cases of GEFS+ (see preceding section). Two mutations are present in families with seizures beginning at 1 to 3 months of age and ending at around 4 months (“neonatal-infantile”) (Heron et al., 2002). Both mutations occur in cytoplasmic loops, inhibiting channel inactivation. A third mutation, also affecting a cytoplasmic loop, was identified in a family with similar seizures beginning at 4 to 12 months of age (“infantile”) (Striano et al., 2006). Of note, ATP1A2 mutations account for some cases of benign familial infantile seizures (see previous section on Familial Hemiplegic Migraine).

Juvenile Myoclonic Epilepsy

Juvenile myoclonic epilepsy accounts for 4% to 10% of all epilepsy (Jallon and Latour, 2005). Featuring myoclonic jerks, generalized tonic-clonic seizures, and absence spells, JME typically begins during adolescence (see Chapter 67). A genetic pattern is clear, but the mixed inheritance pattern suggests multiple heritable causes. One French-Canadian family suffers an autosomal dominant form of JME associated with a mutation in the GABA_A receptor α₁-subunit gene, GABRA1 (Cossette et al., 2002), that results in loss of function and possible neuronal hyperexcitability (Krampfl et al., 2005). Less well described is the possible role of the VGCC β₄-subunit gene, CACNB4, in JME and episodic ataxia (see previous discussion) (Escayg et al., 2000a). GABRD variations may influence susceptibility to JME (Dibbens et al., 2004).

EFHC1 is a gene on chromosome 6p, and five missense mutations have been identified in 6 of 44 JME families in whom it was sequenced (Suzuki et al., 2004). Although the precise function of the EFHC1 protein is unknown, it binds specifically to R-type VGCCs, alters their function, and perhaps thereby influences cell death pathways when transfected into cells in vitro. Whether or not this model turns out to be correct, it illustrates that an inherited channelopathy may result not only from mutation of a channel gene itself, but also of genes whose products regulate the function of otherwise normal channels (Fig. 64.11).

Fig. 64.11 Normal ion channel function relies not only on a normal channel protein. Illustrated are examples of other processes that when defective may theoretically result in aberrant ion channel function and disease.

Chromosome 15q has genetic loci implicated by linkage studies as possible contributors to some cases of JME. This area is known to contain, among other genes, the α₇-subunit of the nicotinic acetylcholine receptor, CHRNA7 (Elmslie et al., 1997).

Similar to JME is familial adult myoclonic epilepsy (FAME) and autosomal dominant cortical myoclonus and epilepsy (ADCME), characterized by autosomal dominant inheritance, adult onset, varying degrees of myoclonus in the limbs, rare tonic-clonic seizures, and a benign course. These syndromes bear some similarity to JME except for the adult onset and the highly penetrant autosomal dominant transmission. Whereas the genes are not yet known, the FAME gene has been mapped to chromosome 8q24 (Plaster et al., 1999) and the ADCME gene to chromosome 2q11.1–q12.2 (Guerrini et al., 2001).

Childhood Absence Epilepsy

Childhood absence epilepsy, a syndrome less common than JME and typified by brief, frequent absence spells (see Chapter 67), exhibits a multigenic pattern of inheritance. Mutations in GABRB3, encoding the GABA_A receptor β₃-subunit, appear to cause CAE in some families (Tanaka et al., 2008; Urak et al., 2006). A mutation in the GABA_A receptor γ₂-subunit gene, GABRG2, on chromosome 5q, conferring loss of benzodiazepine-induced enhancement of GABA-induced currents in vitro, is present in a family with CAE and an increased incidence of febrile seizures (Wallace et al., 2001). Mutations in the T-type VGCC gene, CACNA1H, may contribute to some cases of CAE (Chen et al., 2003) and influence susceptibility to other IGEs (Heron et al., 2007). Mutations in other genes—CACNA1A, GABRA1, and a locus on chromosome 8q24—may also play a role.

Epilepsy and Paroxysmal Dyskinesia

A recognized association between IGE and paroxysmal dyskinesia exists. Characterizing this group of episodic movement disorders are recurrent attacks of dystonia, chorea, athetosis, or ballismus. In a family with an autosomal dominant pattern of generalized epilepsy (mostly absence attacks) and paroxysmal nonkinesigenic dyskinesia (PNKD, a type of paroxysmal dyskinesia in which alcohol, coffee, stress, and fatigue trigger attacks), a mutation of KCNMA1 on chromosome 10q was identified (Du et al., 2005). This gene encodes the α-subunit of the BK channel, a potassium channel that normally activates with both membrane depolarization and a rise in intracellular calcium; the mutation heightens calcium sensitivity, enhancing channel activity. Ethanol may also activate the BK channel (Davies et al., 2003), suggesting a mechanism whereby alcohol consumption in synergy with the mutation may precipitate a dyskinetic attack (Du et al., 2005). That a single ion channel mutation can cause both paroxysmal dyskinesia and epilepsy implies a pathophysiological connection between the basal ganglia and the cortex. PNKD without epilepsy is linked to mutation of the myofibrillogenesis regulator gene, MR1 (Lee et al., 2004b); the mechanism, and whether it involves an ion channel, is unknown.