Clinical manifestations

Presentation and course

The sine qua non for diagnosis of most of these channelopathies is recurrent attacks of weakness (Table 1). Typically, the weakness is generalized and involves the arms and legs together, sparing bulbar and respiratory muscles. Focal weakness of isolated muscles has been described. During attacks, affected muscles are in a state of sustained depolarization and are electrically inexcitable (26). This is reflected clinically in weakness, hypotonia, and areflexia. Attacks may last minutes, hours, or even days. Episodes are often precipitated by specific triggers (behavior/diet) or are associated with abnormal serum potassium levels.

Measurement of serum potassium levels, recognition of precipitating factors, as well as clinical features of myotonia, cardiac arrhythmia, or distinctive craniofacial and skeletal features help in distinguishing variant forms of inherited periodic paralysis.

In the related myotonic disorders without periodic paralysis, myotonic stiffness is the major symptom and is typically either improved with exercise (myotonia congenita) or worsened by exercise (paradoxical myotonia of paramyotonia congenita).

Table 1. Clinical Features of Periodic Paralysis

*weakness in patients with hypokalemic periodic paralysis due to sodium channel mutations worsens with acetazolamide (see text).

Hypokalemic periodic paralysis (hypoPP). In hypokalemic periodic paralysis, episodes of weakness are associated with a decrease in serum potassium levels, sometimes dramatically (eg, less than 2.0 mEq/L). Inheritance is autosomal dominant, typically with complete penetrance, although penetrance may be decreased in women (28). Approximately 60% of patients have mutations in the gene for alpha-subunit of the dihydropyridine-binding voltage-sensitive L-type calcium channel CACNL1A3 (also known as CACNA1S) and are designated hypokalemic periodic paralysis type 1. Ten percent of families not linked to the CACNL1A3 locus have mutations in the gene for skeletal muscle sodium channel SCN4A, which are designated as hypokalemic periodic paralysis type 2. A third group of otherwise typical periodic paralysis comprises patients who do not demonstrate linkage to either the CACNL1A3 or SCN4A loci.

Table 2. Genes detected in HypoPP

*Commercially available in the United States

Familial or primary hypokalemic periodic paralysis symptoms usually begin around puberty. Men are more severely affected than women. Attacks may be triggered by rest after intense exercise, carbohydrate-rich meals, insulin, high-sodium diet, emotional stress, alcohol consumption, or rarely, exposure to cold. A typical attack may last for hours or, less often, days. The attacks may occur repeatedly: daily, weekly, monthly, or less often. Episodes of weakness are not accompanied by sensory, cardiac, or cognitive symptoms. Slowly progressive and chronic proximal leg weakness in the fourth or fifth decades is typical. Serum CK is usually normal, but may be slightly elevated. Muscle biopsy features include typical myopathic features of abnormal increase in central nuclei and variation of fiber size, and vacuoles. Myotonia does not typically occur in hypokalemic periodic paralysis. Patients with sodium channel mutations (hypoPP type 2) may exhibit the following distinctive features: (1) earlier age of onset, (2) postictal myalgia, (3) worsening of symptoms with acetazolamide, (4) prominent tubular aggregates rather than vacuoles on muscle biopsy as seen in type 1, and (5) complete penetrance in males and females (87).

Hypokalemic periodic paralysis also exists in nonfamilial form with predilection for Asian people. Attacks consist of sporadic periodic paralysis and, less commonly, thyrotoxic periodic paralysis. In contrast with familial periodic paralysis, the initial episode of thyrotoxic periodic paralysis typically occurs in persons 20 to 40 years of age and predominantly in males. The clinical manifestations of thyrotoxic periodic paralysis and hypokalemia are indistinguishable from those of familial hypoPP except for elevated blood levels of thyroid hormones and depressed thyroid-stimulating hormone of other etiologies in thyrotoxic periodic paralysis and resolving of symptoms with effective treatment of thyrotoxicosis (51; 78). Typical symptoms and signs of hyperthyroidism may be subtle or absent. Although spontaneous resolution of attacks occurs in a few hours to 2 days, even without potassium chloride supplementation, cardiac arrhythmias and respiratory failure are possible life-threatening complications. The vacuolar myopathy and permanent residual weakness that develop after repeated attacks of familial periodic paralysis are uncommon in thyrotoxic periodic paralysis. The pathogenic gene mutations in nonfamilial periodic paralysis are largely unknown. A mutation in KCNJ18 encodes a skeletal muscle-specific inwardly rectifying K+ channel Kir2.6 in some patients with thyrotoxic periodic paralysis (78) and also in sporadic periodic paralysis (22). The mutation results in amino acid substitution in Kir2.6, which exerts negative inhibition on wild-type Kir2.6 and Kir2.1, another Kir channel expressed in the skeletal muscle. They hypothesized that the decreased K+current from hypofunction of Kir2.6 predisposes the sarcolemma to hypokalemic-induced depolarization, which leads to Na+ channel inactivation and inactivation of muscles.

Hyperkalemic periodic paralysis (hyperPP). In this less common form, episodes of weakness are typically associated with elevated serum potassium levels. However, otherwise typical cases have been described with normal or, rarely, even low ictal potassium levels. Inheritance is autosomal dominant. Most patients have mutations in the gene for the skeletal muscle sodium channel SCN4A. Symptoms begin during early childhood. Males and females are equally affected. Attacks may be triggered by rest after intense exercise, immobility, or fasting, and may be aborted by carbohydrate-rich meals. The duration of episodes is usually shorter than attacks in hypokalemic periodic paralysis, typically lasting for minutes to hours. At the onset of an attack, patients may complain of muscle tightness or nonspecific muscle discomfort. Usually, the frequency of attacks diminishes with age. However, attacks are replaced by fixed proximal weakness, as in hypokalemic periodic paralysis. When this happens, muscle biopsy features are similar to those seen in hypokalemic periodic paralysis, including vacuoles. Serum CK is usually normal or mildly elevated, as in hypokalemic periodic paralysis. A clinical feature distinctive to hyperkalemic periodic paralysis is the occasional presence of electrical myotonia, usually without overt limb myotonia, although patients can manifest eyelid myotonia (difficulty opening eyes after sustained forced eye closure) or percussion myotonia of the tongue. These features are typically exacerbated by cold exposure. The presence of myotonic discharges in a patient with otherwise typical periodic paralysis symptoms establishes hyperkalemic periodic paralysis as the likely cause, as this is not seen in hypokalemic periodic paralysis. The mechanism for this membrane hyperexcitability is probably due to defective fast inactivation of sodium channels during action potential generation (73).

Paramyotonia congenita (PMC). This disorder is allelic to hyperkalemic periodic paralysis (ie, due to mutations in different locations in the same gene), and is caused by mutations in the sodium channel SCN4. Myotonia is the major symptom and is dramatically worsened by cold. Unlike typical myotonia, which improves with repeated contractions (commonly referred to as the warm-up phenomenon), paramyotonia worsens with repeated muscle contraction (hence, the term “para”-myotonia or paradoxical myotonia). A minority of patients experience the warm up phenomenon. Excessive cooling may lead to muscle depolarization and paralysis that may take hours to reverse on warming, without alteration of serum potassium. Serum CK may be elevated up to 10 times normal.

Myotonia congenita (MC). Myotonia congenita may be either autosomal dominant (Thomsen) or recessive (Becker), the latter form being the more common. These 2 conditions are distinguished by the mode of inheritance and the severity of the symptoms. All patients identified thus far have mutations in skeletal muscle chloride channel CLC-1. In both forms, the myotonia is prominent but, unlike PMC, improves with repeated contractions (“warm up”). A minority of patients actually experience paramyotonia, however. In general, the clinical features of both forms are similar, but there are some differences. Myotonia in the recessive form is more severe and disabling, although the phenotype is variable even within affected family members. This type usually presents between 4 and 12 years of age, later than the dominant form that typically presents within the first few years of life with myotonia and muscle stiffness. The severity of weakness is typically worse in the recessive form. Mild fixed proximal weakness is apparent in the recessive from, whereas the dominant form tends to have normal strength, though some individuals eventually develop mild proximal weakness later in life. Episodes of muscle paralysis usually do not occur. A dramatic increase in the bulk and visible definition of voluntary muscles is common in patients with myotonia congenita, presumably because the skeletal muscles are in an almost continuous state of muscle contraction. Systemic complications such as cardiopathy or ventilatory muscle weakness are not seen. There may be an increased risk of malignant hyperthermia with anesthetic agents.

Other myotonic disorders of sodium channels. At least 4 myotonic disorders are associated with sodium channel mutations but lack attacks of paralysis: myotonia fluctuans (72), myotonia permanens (49), chronic myotonia (49), and acetazolamide responsive myotonia (68). Whether these are truly distinct entities or a phenotypic spectrum of the same disorder is not clear. Myotonia fluctuans is characterized by variably symptomatic myotonia without cold sensitivity or weakness; administering potassium can cause severe muscle stiffness (71). Patients with myotonia permanens have unvarying, severe myotonia (49). In acetazolamide responsive myotonia, potassium loading, fasting, and cold exposure result in often painful myotonic stiffness. Given the overlap, the term "sodium channel myotonia" has been suggested for patients with isolated myotonia who have mutations in SCN4A (75).

Andersen-Tawil syndrome (ATS). Andersen-Tawil syndrome is an autosomal dominant disorder, and an uncommon cause of periodic paralysis (accounting for approximately 10% of all periodic paralysis cases). It is characterized by a triad: periodic paralysis, distinctive craniofacial and skeletal anomalies, and prolonged QT interval with a propensity toward malignant ventricular dysrhythmias. However, affected individuals may express only 1 or 2 of the 3 components (94; 80; 65).

Although the episodic weakness was initially thought to be potassium sensitive, the syndrome has been reported in patients with normokalemia, hyperkalemia, and hypokalemia (80). The dysmorphic features are highly variable and include short stature, scoliosis, clinodactyly (permanent lateral or medial curvature of a finger or toe), hypertelorism (wide-set eyes), small or prominent ears that are low-set or slanted, micrognathia (small chin), broad forehead, and dental abnormalities (eg, delayed tooth eruption or missing teeth) (103). Cardiac manifestations vary from an asymptomatic long QT syndrome to life-threatening ventricular tachyarrhythmia requiring an implantable defibrillator, but most patients exhibit either a long QTc interval or a prolonged QU interval with characteristic U-wave morphology (106; 103). Episodes of muscle weakness may fluctuate in severity and usually begin before the age of 10 years old or in adolescence. Mild permanent weakness may be seen in some patients. Muscle creatine kinase levels remain normal or slightly elevated. Muscle biopsy may reveal tubular aggregates, variability of fiber size, and central nuclei. Myotonia is not a common feature in Andersen-Tawil syndrome, but lingual myotonia has been described in 2 patients with hyperkalemic periodic paralysis and cardiac arrhythmia (52; 37). Yoon and colleagues described a distinct neurocognitive phenotype associated with ATS, characterized by deficits in abstract reasoning and executive dysfunction (104). Most patients (approximately 70%) have mutations in KCNJ2 on chromosome 17q23, the gene encoding the inwardly rectifying Kir2.1, a potassium channel expressed in cardiac and skeletal muscle, and brain (65). These cases are designated Andersen-Tawil syndrome type 1. The remaining 30% of patients presumably result from some other as-yet-unidentified mutation, and are designated Andersen-Tawil syndrome type 2.

Prognosis and complications

The long-term prognosis for these diseases is generally good. Hypokalemic paralytic crises tend to be more severe in men just after puberty and in early adult years; over subsequent decades the frequency and severity of attacks tend to remit, although in some they may be replaced by fixed progressive limb weakness. Whether this weakness is in part preventable with long-term carbonic anhydrase inhibitor therapy is uncertain. This progressive myopathy is also seen in some patients with Hyperkalemic periodic paralysis. Although most patients with periodic paralysis enjoy a normal lifespan, complications due to profound weakness or cardiac arrhythmias triggered by abnormal serum levels of potassium during the episodes of weakness can be life threatening if not attended promptly (15). Brancati and colleagues (13) described the first instance of an unusually severe hyperkalemic periodic paralysis or paramyotonia congenita in an Italian family with 9 patients harboring T704M mutation in SCN4A gene. All affected persons had paralytic episodes in the first year of life persisting in adult life with increased frequency, duration, and severity. The paralytic attacks were not responsive to acetazolamide, chlorothiazide, or salbutamol. Furthermore, patients with an unusually severe form of potassium-aggravated myotonia (PAM), myotonia fluctuans, may develop severe myotonia involving intercostal muscles that can interfere with respiratory function. Patients with this and other forms of potassium-aggravated myotonia seem unusually susceptible to depolarizing neuromuscular blockade in worsening myotonia, and these agents should be used with caution. For Andersen-Tawil syndrome, despite the often dramatic ventricular arrhythmias noted on Holter monitoring, the long-term risk, though not well quantified, appears less than that for other forms of long-QT syndrome.

Clinical vignette

A 56-year-old college professor had attacks of episodic weakness that began at the age of 8 years old. The first bout occurred following a vigorous fist fight and lasted 12 hours. Subsequently, episodes of weakness occurred after long car rides, on waking in the morning, after large carbohydrate-rich meals, or with rest after physical exertion. Some episodes were mild, lasting several hours, and others resulted in severe limb paralysis. However, even with profound quadriparesis, speech, swallowing, and respiration were unaffected. His mother had similar episodes. The diagnosis of familial hypokalemic periodic paralysis was made with the documentation of low serum potassium levels (2.5 mEq/L) during spontaneous attacks and with attacks induced by glucose loading. A muscle biopsy at 38 years of age showed many fibers with large vacuoles. Treatment with potassium supplementation blunted the severity of the attacks. With acetazolamide therapy, the attacks ceased for several years, but by the age of 43 years old, despite continued use of acetazolamide, the episodic weakness resumed with less-severe attacks. He also noted fixed progressive leg weakness.

Biological basis

Etiology and pathogenesis

Much of the recent progress in these disorders is the result of expression cloning and voltage-clamp studies in animal systems. For each disorder, defects in either calcium channels (hypokalemic periodic paralysis), potassium channels (Andersen-Tawil syndrome), or sodium channels (hyperkalemic periodic paralysis and a small proportion of hypokalemic periodic paralysis) have been established as the cause; however, the exact mechanisms by which ion channel defects lead to each phenotype remain to be fully characterized. In general, the unifying feature of all skeletal muscle channelopathies is an abnormal and sustained depolarization with electrical inexcitability due to inactivation of sodium channels (periodic paralysis), or partial-membrane depolarization alone or in combination with abnormal and early recovery from sodium channel inactivation that causes recurrent action potentials (myotonic disorders). Although this is a useful distinction, there is considerable overlap: some patients with otherwise typical periodic paralysis demonstrate myotonia, and others with typical myotonic symptoms also experience episodes of paralysis.

Hypokalemic periodic paralysis. Early studies demonstrated an intracellular shift of potassium during episodes of paralysis (107), and subsequent studies on muscle biopsies from patients with hypokalemic periodic paralysis demonstrated a paradoxical membrane depolarization in response to hypokalemia that is potentiated by insulin. That response is not inhibited by tetrodotoxin (a sodium channel blocker), indicating that this depolarization is not mediated by fast-inward sodium current (74).

Most cases of hypokalemic periodic paralysis are genetically linked to the CACNA1S gene on chromosome 1q31-32, a region coding for the dihydropyridine-sensitive calcium channel CaV1.1, located in the sarcoplasmic reticulum and responsible for initiating muscle contraction (31). Previously, all known mutations were noted to replace a positively charged arginine with a neutral amino acid, and all are in the voltage-sensitive membrane spanning segment (S4) of domains II, III, or IV of the alpha subunit of CaV1.1 (41; 67; 21). How mutations in CACNA1S cause an abnormal membrane response to hypokalemia is not clear, as functional studies suggested that mutations in CACNA1S do not significantly alter L-type calcium currents in cultured myotubes from patients with hypokalemic periodic paralysis. Skeletal muscle fibers in hypokalemic periodic paralysis are susceptible to depolarization-induced inexcitability, and the membrane abnormalities may contribute the genesis of pathological depolarization and depolarization-induced paralysis. Ruff and colleagues demonstrated that insulin potentiates depolarization of hypokalemic periodic paralysis fibers in vitro by reducing inward rectifier K+ conductance, and the Ca2+ channel mutations indirectly derange membrane excitability by altering the function of other membrane channels (76). A study confirmed the reduced inward rectifier K+ currents in muscles fibers of patients with hypokalemic periodic paralysis (69). Other electrophysiologic studies in hypokalemic periodic paralysis suggest enhanced sodium channel inactivation and reduced current (42) or impairment of skeletal muscle ATP-sensitive K+ channel activity (99). This is further supported by the finding that acetazolamide and other carbonic anhydrase inhibitors known to be effective in hypokalemic periodic paralysis open BK-type Ca-activated K+ channels (97).

However, the situation is further complicated because hypokalemic periodic paralysis is also caused by similar mutations in the SCN4A gene coding for domains II and III of voltage sensor S4 of the alpha subunit in the skeletal muscle sodium channel NaV1.4 (hypoPP2), the same gene that is usually associated with hyperPP (14; 42; 09; 87). These mutations are thought to cause an enhanced slow inactivation of NaV1.4 channels making them less available to excitation, leading to muscle membrane hypoexcitability, and impairing the ability to recover from fast- or slow-inactivation states (19; 90). Unitary sodium conductance does not appear to play a role in hypoPP2 (19). Electrophysiological studies of rat NaV1.4 channels with the R663H mutation (human R669H ortholog) suggest that inward positioning of the voltage sensor created by this mutation leads to exposure of a permeability pore that allows inward flux of protons, thus, potentially altering intracellular pH and possibly contributing to the altered muscle membrane physiology ultimately associated with paralytic episodes (84; 90). Additional studies have determined that this pore may also allow sodium ion influx as well, depending on which amino acid is substituted for arginine at these locations (91; 56). Another study of rat NaV1.4 channels missense mutation orthologous to the human hypoPP mutations shows an anomalous gating pore current whereas the homologous mutation in the domain IV voltage sensor in NaV1.4 associated with paramyotonia congenita does not (34). This small current may cause the aberrant depolarization during the paralysis attacks.

Yet another line of evidence suggesting a role of disorders of channel trafficking comes from a study of myotubes from muscle derived from hypokalemic periodic paralysis patients with CACNA1S mutations indicating downregulation of channel expression. An important and largely unresolved issue is how SCN4A and CACNA1S mutations result in inexcitability of the muscle membrane in the presence of hypokalemia (48).

Moreover, the DHP-receptor calcium channel encoded by CACNA1S associates with the ryanodine receptor (RyR1) to form the calcium release channel that is the basis for excitation-contraction coupling in skeletal muscle (67). Most patients with malignant hyperthermia (a disease characterized by abnormal release of calcium from the sarcoplasmic reticulum during anesthesia) have mutations in RyR1. However, a separate malignant hyperthermia locus was identified on chromosome 1q, and mutations in CACNA1S have now been identified in certain kindreds as a less common cause of malignant hyperthermia susceptibility. Malignant hyperthermia and hypokalemic periodic paralysis may, therefore, be regarded as allelic disorders of the DHP-sensitive calcium channel. Interestingly, unlike hypokalemic periodic paralysis, in which mutations in CACNA1S occur in the voltage sensing domains, malignant hyperthermia mutations occur in the intracytoplasmic loop between domains II and III, a portion of the molecule which interacts with the ryanodine receptor (60).

Finally, Abbott and colleagues described a missense mutation (R83H) in KCNE3, the gene coding for MinK-related peptide 2 (MiRP2) in 2 families, 1 with hypoPP and another with hyperkalemic periodic paralysis (hyperPP) (01). MiRP2 is an accessory subunit that coassembles with the alpha subunit of the voltage-gated K+ channel, Kv3.4, which contributes to resting membrane potential in skeletal muscle. A confirmatory study, however, failed to confirm the KCNE3 mutation as a disease-causing mutation as opposed to a benign polymorphism, and a disease-causing mutation in this third group of patients remains to be proven (88; 40). It is uncertain whether this is a mutation or a benign polymorphism. An abnormal and pH-dependent reduction in inward ionic current in R83H MiRP2-K3.4 channels has been expressed in Chinese hamster ovary cells, a finding that suggests a pathophysiologic role of KNCE3 mutations in periodic paralysis. This may explain the exercise-induced membrane depolarization in periodic paralysis patients and could arise from a reduced K current with intracellular acidosis following prolonged or intense exercise (02).

Hyperkalemic periodic paralysis and sodium channel myotonias. Early physiological data implicated a potassium-induced alteration in membrane sodium conductance in hyperkalemic periodic paralysis. In vitro studies of muscle from patients with hyperkalemic periodic paralysis demonstrated persistent inward current that was potentiated by potassium and blocked by tetrodotoxin (indicating that it is likely mediated by persistent sodium influx) (70). Patch clamp analysis of cultured myotubes from an individual with hyperkalemic periodic paralysis revealed that elevated levels of extracellular potassium induced abnormal sodium channel inactivation, with prolonged or repetitive openings (16). Therefore, it came as little surprise that linkage studies implicated SCN4A, the gene encoding the alpha-subunit of the skeletal muscle voltage-sensitive sodium channel Skm-1, on chromosome 17q as the causal gene for hyperkalemic periodic paralysis as well as paramyotonia congenita (30; 66; 75).

In general, the phenotypes of both hyperkalemic periodic paralysis and paramyotonia congenita can be explained (at least in part) by defective inactivation of voltage-gated sodium channels, alone or in combination with accelerated recovery from inactivation. Summarizing these complex data, 2 regions of the sodium channel are involved in normal inactivation, which may be altered in periodic paralysis or paramyotonia congenita: (1) the intracellular loop between domains III and IV that is thought to act as an inactivation particle, or (2) the docking site that functions as a receptor for the inactivation gate (III-IV loop). Many of the mutations identified occur in locations that interrupt either function; for example, one of the glycine pair (Gly 1306) is substituted in 3 different mutations that sterically limit movement of the III-IV loop, thus, impairing sodium channel inactivation that results in membrane depolarization due to persistent inward sodium current. The exact mechanism by which each mutation causes either a predominantly myotonic phenotype or a paralytic phenotype is probably based on the mutation’s ability to interfere with sodium channel fast inactivation or slow inactivation, causing either slight depolarization that does not inactivate sodium channels but allows recurrent activation (myotonic phenotype) or more severe depolarization that leads to sodium channel inactivation and inexcitability (paralytic phenotype). Slow inactivation of sodium channels is based on conformational changes within the structure of the channel and occurs over a much longer period as compared to fast inactivation (18; 17; 39). Impaired slow sodium inactivation appears to correlate with the hyperkalemic periodic paralysis phenotype, whereas mutations that do not affect slow inactivation but interfere with fast inactivation are associated with nonparalytic, potassium-sensitive myotonic disorders (39; 101). The correlation is, however, incomplete because some hyperkalemic periodic paralysis patients demonstrate electromyographic, and sometimes clinical, myotonia. Furthermore, how hyperkalemia potentiates each phenotype is not clear.

Andersen-Tawil syndrome. Andersen-Tawil syndrome is caused by mutations in KCNJ2, the gene encoding the alpha subunit of Kir2.1, a member of the inwardly rectifying K+ channel family (65). Andersen-Tawil syndrome is the first human disease recognized as being caused by mutations in Kir channels. To date, over 30 mutations have been identified in KCNJ2. Kir2.1 channels are expressed predominantly in heart, brain, and skeletal muscle, and mutations are associated with: ATS, atrial fibrillation, and a short QT syndrome (06). Of the more than 30 mutations, the majority of these result in a variable dominant negative effect on the Kir2.1 channel. These channels are constructed and trafficked properly to the membrane either alone or coassembled as hetero-tetramers with wild-type channel subunits (11; 53). Once at the membrane, they malfunction depending on the site of the mutation. A variety of mechanisms, including PIP2 (phosphatidylinositol 4,5-bisphosphate) binding interference, exaggerated inhibition by intracellular magnesium ions, altered gating, and protein misfolding have been demonstrated (07; 54; 93). The variability of this effect may provide a molecular basis for genotype-phenotype mismatch, variable penetrance, and expressivity of ATS phenotypes (92). Three or four of the known mutations appear to interfere with channel trafficking and membrane placement as demonstrated by confocal microscopy and immunofluorescence techniques (06). Lastly, proteins and cofactors associated with trafficking Kir2.1 channels from the endoplasmic reticulum have also been implicated in ATS. GRIF-1 (γ-aminobutyric acid receptor-interacting factor 1), the TRAK family of proteins, and O-GlcNAc transferase are just a few potential candidates for proteins with mutations that may explain the remaining 30% of ATS patients without KCNJ2 mutations (38).

The function of Kir2.1 in cardiac muscle has been partly characterized. Kir2.1 channels are responsible for the major part of the cardiac current IK1 during the terminal repolarization phase of the cardiac action potential. Strong inward rectification prevents excess loss of K+ during the plateau phase of the cardiac action potential. Reduction of Kir2.1 function would be expected to prolong the duration of cardiac action potential and QT interval by reducing the amount of repolarizing current during the terminal phase. Action potential prolongation is a prerequisite for early afterdepolarizations, a potential trigger for ventricular tachyarrhythmias. These afterdepolarizations have been shown to be suppressed by hyperkalemia and facilitated by hypokalemia and catecholamines or sympathetic stimulation (92; 100; 61).

Bendahhou and colleagues also demonstrated impaired skeletal muscle membrane excitability in individuals with Andersen-Tawil syndrome, similar to that seen in hypokalemic periodic paralysis. However, the mechanism by which this occurs remains unclear (11). Possible mechanisms include sodium channel inactivation through persistent membrane depolarization, which may arise from either reduction in inwardly rectifying current, trapping of potassium in transverse tubules, or both.

Kir2.1 channels are also expressed in the developing fetus. Kir2.1 knockout mice have a complete cleft of the secondary palate and a slight narrowing of maxilla (85; 105), a potential analogy to facial dysmorphism seen in patients with Andersen-Tawil syndrome; however, the exact role of Kir2.1 in human embryonic development is unclear.

Myotonia congenita (MC). Several physiological studies in the last 20 years have suggested that chloride conductance is reduced in mouse and goat models of myotonia congenita (04). Linkage studies in myotonia congenita have shown that both the dominant (Thomsen disease) and the recessive generalized form (Becker type) are linked to a locus on chromosome 7 that encodes CLC-1 chloride channel (03; 36). At least 30 mutations in the chloride channel have, thus far, been identified (48; 64). Normally, chloride conductance is moderate at the resting membrane potential, increased with depolarization, and reduced with hyperpolarization. Dominantly inherited mutations interfere with voltage dependency of chloride channel function. For example, in the case of the dominant G200R mutation, a marked depolarization shift results in closure of chloride channels that are normally open at the resting membrane potential. The dominant negative effect is explained by coassembly of mutant channels with normal wild-type channels. Resultant chloride conductance is dramatically reduced despite normal channel expression, and myotonia occurs (86; 08). Other experimental evidence suggests that chloride conductance needs to fall to at least 25% of normal before myotonia occurs, thus, explaining the requirement for loss of both function alleles in recessive mutations that generally result in expression of nonfunctional channel protein. Other reports suggest the presence of chloride channel mutations that defy classic Mendelian inheritance patterns; the same mutation may be either autosomal recessive or dominant with reduced penetrance or incomplete dominance in certain kindreds (45; 64). This is exemplified by the observation that electromyographic myotonia can be found in about two thirds of clinically unaffected parents of recessively inherited myotonia congenita (24).

Epidemiology

The prevalence of these disorders is not well defined. In general, these disorders are rare, probably affecting no more than 4 to 5 per 100,000 in the population. The prevalence of hypokalemic paralysis may be as low as 0.4 per 100,000 (43).

Prevention

Some triggering factors are well established for some of the conditions, and their avoidance can help minimize symptoms. In myotonic patients, cold often accentuates muscle stiffness and, in patients with paramyotonia congenita, may trigger episodes of paralysis. Hyperkalemic attacks are triggered by hypoglycemia and potassium intake (and improved with carbohydrate ingestion), whereas hypokalemic attacks may be induced by carbohydrate ingestion (and improved with potassium intake).

In ATS patients, medications that may prolong QTc should be avoided in order to prevent life-threatening ventricular arrhythmias or sudden cardiac death (63). Although not reported in typical hypoPP patients, corticosteroids have been implicated in provoking paralytic episodes in ATS patients. This is thought to be due to either the hypokalemic effect via mineralocorticoid activity or direct effects on K⁺ channel activity (10).

Differential diagnosis

In most cases, the combination of recurrent episodic weakness and documentation of abnormal ictal serum potassium levels, with or without a positive family history rarely pose challenges for diagnosis. However, when variations in serum potassium levels remain within the normal range, or if the weakness seems less clearly paroxysmal in nature, the diagnosis can be challenging. Generally, the differential diagnosis includes secondary causes of periodic paralysis that include hypokalemia or hyperkalemia from renal, adrenal, gastrointestinal, and drug-induced causes as well as thyrotoxicosis. Thyrotoxic periodic paralysis appears more commonly in Asians, suggesting an underlying genetic predisposition. At least 1 case report of hypoPP caused by Gitelman syndrome, an autosomal recessive renal tubular disorder typically characterized by mild to moderate hypokalemia, metabolic alkalosis, hypomagnesemia, and hypocalciuria has been documented. Occasionally, myasthenia gravis and other neuromuscular transmission disorders may be confused with periodic paralysis. However, ocular and bulbar weakness is typical of myasthenia and rarely, if ever, occurs in periodic paralysis. Rarely, a cocaine “binge” can induce acute onset hypokalemic paralysis (62). In these cases, hypokalemia may be induced by intracellular shifts of K+ secondary to the adrenergic effects of cocaine. Excessive glycyrrhizic acid intake in the form of licorice may cause secondary hypokalemic periodic paralysis through induction of a hypermineralocorticoid state.

For the nondystrophic myotonias (eg, myotonia congenita, paramyotonia congenital, and potassium aggravated myotonia), the differential includes one of the myotonic myopathies, myotonic dystrophy (DM1), or proximal myotonic myopathy (PROMM, DM2) (96). The rare childhood disorders Schwartz-Jampel syndrome (caused by mutations in the basement membrane protein perlecan) and Brody disease (caused by mutations in ATP2A1 leading to abnormal muscle contractures with exercise: “pseudo-myotonia”) can also be mistaken for one of the nondystophic myotonic disorders.

Diagnostic workup

The diagnosis of periodic paralysis can often be confirmed by genetic testing. This is recommended as the initial diagnostic step (after a careful history and exam) when there is adequate clinical suspicion. This basis of this clinical suspicion is arrived by establishing a characteristic history, presence of abnormal serum potassium levels during spontaneous attacks, and presence of characteristic abnormalities on electromyography (myotonia) and long or short exercise nerve conduction test. In the 30% of patients with periodic paralysis that do not yet have an identified genetic mutation, further investigation should be performed to exclude secondary forms of periodic paralysis such as thyrotoxicosis or other causes of abnormal blood potassium level (82).

Electrodiagnostic evaluation of patients with a primary periodic paralysis is usually performed using either the short (89) or long exercise test (58). The overall sensitivity of the latter test has been estimated at 60% to 80% with high specificity, using a cutoff of a 40% decline in CMAP amplitude or area, following 5 minutes of exercise. The exercise nerve conduction study can also be useful in documenting the presence of a muscle membrane abnormality when the history is vague or atypical (46). Fournier and colleagues distinguished the mutation type based on a battery of electrodiagnostic tests with short and long exercise tests and the presence of after-discharges and myotonia on electromyography, with good sensitivity and specificity (32). For the long exercise CMAP test, compound muscle action potentials (CMAPs) are recorded from the abductor digiti minimi muscle, supramaximally stimulating the ulnar nerve at the wrist. The subject rests for 10 minutes prior to the test, then CMAP amplitude and area are recorded every minute for 5 minutes to establish a stable baseline. The fingers are then abducted against resistance for 5 minutes, with brief 3 to 4-second relaxations to avoid ischemia. This is followed by a 40 to 60-minute postexercise rest period. Supramaximal stimulations are administered every minute during exercise and every 2 minutes during the postexercise period. A reduction in CMAP amplitude of 40% or more is considered abnormal and is reportedly seen in more than 70% of patients (82).

Characteristic electrodiagnostic findings along with a clinical history can be helpful in determining which genetic mutations are most likely. The presence of clinical or electrical myotonia with a characteristic history is virtually diagnostic of hyperkalemic periodic paralysis due to a sodium channel mutation. Attention should be given to presence of craniofacial and skeletal anomalies that would suggest the diagnosis of Andersen-Tawil syndrome. An EKG should be performed in all patients to identify repolarization abnormalities such as QTc prolongation or prominent U wave suggestive of Andersen-Tawil syndrome. When potassium levels cannot be obtained during spontaneous attacks of weakness, provocative tests in the form of glucose or potassium loading have proven helpful in anecdotal experience. These provocative tests should be performed with caution in the hospital with continuous cardiac monitoring because of possible exacerbation of weakness (and cardiac arrhythmias). To date, guidelines for these challenges have not been established and are the subject of ongoing research. The work by Fournier and colleagues suggests that although the short exercise test may not be helpful in distinguishing between hypoPP and hyperPP, the long exercise test might help differentiate between the 2 conditions. They demonstrate that patients with hyperPP are more likely to have a significant increment in CMAP amplitude immediately after exercise whereas hypoPP will not. Both will have the characteristic significant decrement in CMAP (32). Muscle biopsy is usually not required, although the presence of vacuolar changes or tubular aggregates can provide further suggestive evidence of periodic paralysis when the history is atypical. Diagnosis by mutation analysis in sodium, calcium, and potassium channel genes is available at specialist centers throughout the United States and Europe but is not routinely employed in many institutions, especially in settings where the clinical history is strongly suggestive. The role of genetic testing is likely to significantly expand in the future.

Management

Management of acute attacks of periodic paralysis is directed at patient education and lifestyle changes to minimize triggers of periodic paralysis, correcting abnormal potassium levels, and use of carbonic anhydrase inhibitors. In some patients, diet alterations and behavioral strategies for acute attacks can be beneficial in decreasing the number or severity of attacks.

Pharmacologic interventions for periodic paralysis are limited. Carbonic anhydrase inhibitors (acetazolamide and dichlorphenamide) have been empirically used for decades. The short and long term effects of dichlorphenamide on attack frequency and quality of life in hyperkalemic and hypokalemic periodic paralysis were studied with 2 multicentered, randomized double-blind, placebo-controlled trials lasting 9 weeks followed by a 1 year extension phase (79). The dose for dichlorphenamide in treatment-naïve patients was 50 mg twice daily, whereas patients already taking acetazolamide were assigned an equivalent dosage of dichlorphenamide, calculated as 20% of the acetazolamide dosage. In the hypokalemic periodic paralysis trial, the median attack rate, severity-weighted attack rate, and attack duration were significantly reduced with dichlorphenamide. In the hyperkalemic periodic paralysis trial, the median severity-weighted attack rate was significantly reduced in the dichlorphenamide group compared to placebo, although the group differences in median weekly attack rate and duration did not reach significance. The most common side effects were paresthesias, cognitive disorders, dysgeusia, and renal calculi. The maximum recommended dose is 200 mg daily.

There are no randomized, controlled trials performed with acetazolamide in periodic paralysis, although it is routinely used. Its use is largely based on anecdotal reports and nonrandomized, single-blind trials (82). There is still lack of sufficient evidence for a full guideline of standardized regimen on when to start the treatment. In both hypokalemic periodic paralysis and hyperkalemic periodic paralysis, acetazolamide (typically 250 mg 2 or 3 times a day) is effective in ameliorating or reducing the frequency and severity of attacks. Another agent, pinacidil, a potassium channel opener, also improved muscle strength in 2 of 4 patients in a placebo-controlled double blind study (50).

Investigations into the possible mechanisms of acetazolamide suggest that it enhances resting chloride conductance by inhibiting carbonic anhydrase, an enzyme that is important in regulating intracellular and extracellular pH in skeletal muscle (25; 77). Chloride conductance is thought to have a stabilizing influence on the membrane potential. This is important in skeletal muscle because K⁺, which accumulates in the T-tubules during the process of repolarization, would induce repeated depolarizations if the membrane potential was not stabilized by a high resting chloride conductance. Because the paralysis of hypoPP and hyperPP results from a depolarization-triggered reduction or loss of membrane excitability due to Na⁺ channel inactivation, it is thought that this increased stabilization of the resting potential by enhanced Cl^- conductance may actually thwart these pathologic depolarization currents (77). Eguchi and colleagues demonstrated acetazolamide’s ability to acidify the intracellular environment in cultured human embryo kidney cells. This acidification indirectly shifts Cl^- channel function into an open state (25). Although it is known that human skeletal muscle has a different arrangement of carbonic anhydrase isoenzymes, these studies may provide an understanding of how acetazolamide exerts its effects on chloride channels resulting in membrane stabilization (25). Further evidence suggests there may also be an additional effect on Ca²⁺-activated K⁺ channels, increasing K⁺ currents, and providing an extra membrane stabilizing force (98).

Paradoxically, acetazolamide may increase the frequency and severity of attacks in patients with hypokalemic periodic paralysis due to sodium channel mutations; this is sometimes a clue to the presence of a sodium channel mutation as opposed to the more common calcium channel mutations (09; 87). One cohort study in patients with genetically confirmed hypokalemic periodic paralysis showed 31 of 55 (56%) patients with CACNA1S mutations benefitted from acetazolamide compared with only 3 of 19 (16%) patients with SCN4A mutations (p < 0.002) (57).

For acute hyperkalemic episodes, mild, non-resistive exercise at attack onset and/or a carbohydrate snack may be helpful. Beta-agonists such as salbutamol (1-2 puffs of 0.1mg) and other beta agonists are often effective (82). Acute Hypokalemic attacks often respond simply to oral potassium loading over several hours, although severe crisis may warrant intravenous potassium loading. It is recommended that such intravenous therapy employ mannitol solutions rather than dextrose in water because the carbohydrate load of dextrose may worsen the weakness.

To some degree, hypokalemic attacks may be prevented by reducing total dietary carbohydrates, avoiding large carbohydrate loads, and reducing dietary sodium. If the carbonic anhydrase inhibitors are ineffective or cause unacceptable adverse effects, potassium sparing diuretics such as spironolactone or triamterene may prove useful. Reciprocally, hyperkalemic patients may benefit from daily use of oral diuretics (such as thiazides) that increase urinary potassium. Mild hyperkalemic attacks may be aborted simply by ingestion of oral carbohydrates. Hyperkalemic paralysis patients often carry candy bars or glucose tablets for this purpose. Myotonia in patients with typical hyperkalemic periodic paralysis is usually minimally symptomatic, unless patients have concomitant symptoms of paramyotonia congenita. A remarkably good response is often seen with mexiletine, given to adults at a dose of 200 mg 2 or 3 times a day. Given that myotonia in hyperkalemic paralysis patients is a direct consequence of sodium channel mutations and resulting failure of inactivation, it is not surprising that this drug, a lidocaine analogue with potent, activity-dependent sodium channel blocking properties, should reduce myotonic muscle stiffness. Fortunately, many patients with myotonia congenita also respond favorably to mexiletine.

The episodes of muscle weakness in Andersen-Tawil syndrome respond well to carbonic anhydrase inhibitors like attacks in hypokalemic periodic paralysis or hyperkalemic periodic paralysis; however, the effect on cardiac muscle is not clear. In patients with Anderson-Tawil syndrome, empiric treatment with an antiarrhythmic agent should be considered for significant, frequent, ventricular arrhythmias in individuals affected with reduced left ventricular function (23). In Andersen-Tawil individuals with confirmed KCNJR mutation, flecainide, a type 1c antiarrhythmic, may reduce cardiac arrhythmias (59). Others have also documented benefits of flecainide (33; 81).

Beta-blockers, calcium channel blockers, or amiodarone may also be beneficial in reducing arrhythmias. Some antiarrhythmic drugs (lidocaine, mexiletine, propafenone, quinidine) may worsen attacks of paralysis and should be used with caution in these patients. Salbutamol inhalers should be avoided as they may exacerbate cardiac arrhythmias. Thiazide and other potassium-wasting diuretics should also be avoided as they may induce hypokalemia and aggravate the QT interval (83).

Routine prophylactic implantable defibrillator placement is not recommended; however, any patient who develops sustained ventricular tachycardia, Torsade de pointes, or syncope should be evaluated by a cardiologist for possible defibrillator implantation. The role of routine antiarrhythmic therapy in such patients is controversial and best managed by a cardiologist.

Outcomes

The long-term prognosis for these diseases is generally good. Hypokalemic paralytic crises tend to be more severe in men just after puberty and in early adult years; over subsequent decades the frequency and severity of attacks tend to remit, although in some they may be replaced by fixed progressive limb weakness. Whether this weakness is in part preventable with long-term carbonic anhydrase inhibitor therapy is uncertain. This progressive myopathy is also seen in some patients with Hyperkalemic periodic paralysis. Although most patients with periodic paralysis enjoy a normal lifespan, complications due to profound weakness or cardiac arrhythmias triggered by abnormal serum levels of potassium during the episodes of weakness can be life threatening if not attended promptly (15). Brancati and colleagues (13) described the first instance of an unusually severe hyperkalemic periodic paralysis or paramyotonia congenita in an Italian family with 9 patients harboring T704M mutation in SCN4A gene. All affected persons had paralytic episodes in the first year of life persisting in adult life with increased frequency, duration, and severity. The paralytic attacks were not responsive to acetazolamide, chlorothiazide, or salbutamol. Furthermore, patients with an unusually severe form of potassium-aggravated myotonia (PAM), myotonia fluctuans, may develop severe myotonia involving intercostal muscles that can interfere with respiratory function. Patients with this and other forms of potassium-aggravated myotonia seem unusually susceptible to depolarizing neuromuscular blockade in worsening myotonia, and these agents should be used with caution. For Andersen-Tawil syndrome, despite the often dramatic ventricular arrhythmias noted on Holter monitoring, the long-term risk, though not well quantified, appears less than that for other forms of long-QT syndrome.

Special considerations

Pregnancy

We have encountered women with hypokalemic paralysis whose only attacks were associated with late-term pregnancy and then delivery. Pregnancy does not appear to significantly affect patients with hyperkalemic paralysis. However, there was one case report of a patient with genetically confirmed hyperkalemic periodic paralysis who had marked decrease in the frequency of the paralysis attacks during the first trimester, and the paralysis eventually completely disappeared during the second and third trimesters. In this case, the attacks restarted with the same intensity and frequency after delivery as before pregnancy, but myotonia persisted between the attacks (29). Acetazolamide is contraindicated in pregnancy due to fetal development abnormalities and malformations. There have been anecdotal reports of prevention of paralytic attacks in hyperPP during pregnancy by using salbutamol alone (55). Regional anesthesia is the preferred method during pregnancy.

Anesthesia

There are reports of attacks of hyperkalemic periodic paralysis induced by surgery with general anesthesia. In general, we recommend avoiding depolarizing muscle relaxants. Although the specific factors inciting the attacks are not defined, anesthesia may be safely administered to these patients with appropriate precautions to prevent perioperative potassium excess (eg, minimizing potassium-containing fluid replacement and medications that inhibit potassium excretion), carbohydrate depletion, and perioperative hypothermia, which may aggravate myotonia. Conversely, appropriate potassium repletion and avoidance of carbohydrate load will minimize perioperative paralysis in hypokalemic periodic paralysis patients. It is not clear whether an increased risk of malignant hyperthermia susceptibility exists in periodic paralysis patients; although abnormal caffeine contracture tests have been described in excised muscle from patients with inherited myotonia or periodic paralysis, these tests lack specificity for the diagnosis of malignant hyperthermia susceptibility in this setting (47).