This variant has been associated, so far, with one specific missense mutation (G406R) in the voltage-gated calcium channel gene (CACNA1c).⁹ G406R produces sustained inward Ca²⁺ currents by causing nearly complete loss of voltage-dependent inactivation.⁹ In the heart, prolonged Ca²⁺ current delays cardiomyocyte repolarization and increases the risk of arrhythmia.

Prevalence

In 1975, it was already suggested that LQTS “could be more unrecognized than rare.”¹ However, until now the prevalence was assumed to be anywhere between 1 in 5000 and 1 in 20,000. None of these estimates was based on actual data.

The first data-based assessment of the prevalence of LQTS comes from the largest prospective study of neonatal electrocardiography ever performed, which involved 44,596 infants of 3 to 4 weeks of age in 18 Italian maternity hospitals.¹⁰ An electrocardiogram (ECG) was performed in 44,596 infants 15 to 25 days old (43,080 whites). In infants with a QTc greater than 450 ms, the ECG was repeated within 1 to 2 weeks. Genetic analysis, by screening seven LQTS genes, was performed in 28 (90%) of 31 and in 14 (50%) of 28 infants with a QTc greater than 470 ms or between 461 and 470 ms, respectively, regarded as markedly prolonged by the European Task Force on Neonatal Electrocardiography.¹¹ QTc readings of 451 to 460 ms, 461 to 470 ms, and greater than 470 ms were observed in 184 (0.41%), 28 (0.06%), and 31 (0.07%) infants, respectively (Figure 62-1). Among genotyped infants, disease-causing mutations were found in 12 (43%) of 28 with a QTc greater than 470 ms and in 4 (29%) of 14 with a QTc of 461 to 470 ms. One genotype-negative infant (QTc of 482 ms) was diagnosed to be affected by LQTS on clinical grounds. Among family members of genotype-positive infants, 51% were found to carry disease-causing mutations. In total, 17 of 43,080 white infants were affected by LQTS, which demonstrated a prevalence of at least 1 per 2534 apparently healthy live births (95% confidence interval [CI] 1:1 to 583 to 1:4 to 350). Among them, 1.4% had a QTc between 440 ms and 469 ms and 0.7 of 1000 had a QTc of 470 ms or greater. As 43% of the infants with QTc 470 ms or greater (0.7 per 1000) and 29% of the infants with a QTc between 460 ms and 469 ms have disease-causing mutations and as at least some of the infants with a QTc between 450 ms and 459 ms are also likely to be mutation carriers, it follows that the prevalence of LQTS must be close to 1 per 2000. This does not include the silent mutation carriers (QTc < 440 ms), a group ranging between 10% and 36% according to genotype.⁶ This is the first time that the prevalence of a cardiac disease of genetic origin has been quantified on the basis of actual data.

FIGURE 62-1 Distribution of the 43,080 white neonates among five subgroups (absolute numbers and percentage) according to QTc duration on the screening electrocardiogram. Neonates positive at the genetic analysis are also reported.

(From Schwartz PJ, Stramba-Badiale M, Crotti L, et al: Prevalence of the congenital long-QT syndrome, Circulation 120:1761–1767, 2009.)

Clinical Presentation

The large increase in the number of patients diagnosed as affected has significantly modified the perception of the natural clinical history of the disease. Since the early 1970s, when the first consistent series of patients were reported, the typical clinical presentation of LQTS was considered to be the occurrence of syncope or CA, precipitated by emotional or physical stress, in a young individual with a prolonged Q-T interval on the surface ECG.¹ If these symptomatic patients were left untreated, the syncopal episodes would recur and eventually prove fatal in the majority of cases. This concept was largely based on the fact that the patients initially diagnosed were those most severely affected. It has become progressively evident that this traditional picture represents an oversimplification, and it now appears that even though some patients have the severe manifestations described above, numerous patients have a very benign course as well. Unfortunately, the occasional occurrence of sudden death cannot yet be predicted. Evidence that modifier genes, some of them associated with the release of autonomic mediators, contribute to the highly different severity of the clinical manifestations of LQTS is growing.¹²^–¹⁵

When family screening is performed, prolongation of the Q-T interval can be often detected, and a family history of fainting episodes or of sudden unexpected deaths in an early age is often present. Nonetheless, numerous sporadic cases (approximately 30%), that is, patients with syncope and a prolonged Q-T interval but without clinical evidence of familial involvement, do exist.⁶

The clinical history of repeated episodes of loss of consciousness under emotional or physical stress is typical and unique and is therefore unmistakable; however, the physician must be aware of the existence of LQTS to make the correct diagnosis. However, the clinical presentation is not always clear, so sometimes the diagnosis may be uncertain. The highly malignant J-LN syndrome will not be discussed here ²; it must, however, be noted that its clinical course is very severe and that the patients become symptomatic at a very early age (Figure 62-2). Interested readers are referred to a comprehensive report on this variant.³

FIGURE 62-2 Kaplan-Meier curves of event-free survival. A, all patients with Jervell and Lange-Nielsen (J-LN) syndrome. B, All patients with J-LN, with a magnified view of the events occurring in the first 15 years of life. C, Patients with J-LN versus symptomatic patients with Romano-Ward syndrome of known genotype.¹⁰ D, Patients with J-LN versus symptomatic patients with LQT1 and an unselected LQT1 population.^10,¹²

(From Schwartz PJ, Spazzolini C, Crotti L, et al: The Jervell and Lange-Nielsen syndrome. Natural history, molecular basis, and clinical outcome, Circulation 113:783–790, 2006.)

The two cardinal manifestations of LQTS are syncopal episodes and electrocardiographic abnormalities. The latter have been repeatedly described. Interested readers are referred to previous reviews.⁵ Here we will mention only T wave alternans and a specific echocardiographic pattern.

T-Wave Alternans

In 1975, Schwartz proposed that T-wave alternans represents a characteristic ECG feature of LQTS.¹⁶ Beat-to-beat alternation of the T wave, in polarity or amplitude, may be present at rest for brief moments but usually appears during emotional or physical stress and may precede torsades de pointes (TdP) (Figure 62-3). T-wave alternans is a marker of major electrical instability, and it identifies patients at particularly high risk. Its transient nature limits the possibility of its observation. This is a rather gross phenomenon that should not go unnoticed, when present. The observation of T-wave alternans in a patient with LQTS who is already on therapy strongly suggests the presence of a persisting high degree of cardiac electrical instability, which should prompt reassessment of therapy.

FIGURE 62-3 Five-year-old patient with Jervell and Lange-Nielsen syndrome. Tracing recorded during a syncopal episode. T-wave alternans precedes the onset of torsades de pointes (TdP). TdP is not preceded by a pause.

(From Pernot C: Le syndrome cardio-auditif de Jervell et Lange-Nielsen. Aspects electrocardiographiques, Proc Ass Eur Paediat Cardiol 8:28–36, 1972.)

Echocardiographic Abnormalities

LQTS is still regarded by many as a purely electrical disease that does not involve any mechanical alterations. However, a case-control study demonstrated the frequent presence of highly unusual echocardiographic abnormalities among the patients.¹⁷ These abnormalities included an increased rate of thickening in the early phase of contraction and the presence of a slow movement in the late thickening phase with a plateau morphology sometimes accompanied by a second peak. Recently, these findings have been fully confirmed by a Norwegian study.^18,¹⁹ LQTS is, therefore, not a purely electrical disease. These abnormalities were more frequent in symptomatic patients than in asymptomatic patients, which suggests that these abnormalities may reflect the presence of an arrhythmogenic mechanism. The evidence that the calcium-entry blocker verapamil completely normalizes the contraction pattern suggests that symptomatic LQTS patients may have an abnormal increase in the intracellular calcium concentration before relaxation is completed and that this may be related to an early after-depolarization (EAD): The contraction abnormality would be the mechanical equivalent of an EAD.²⁰ The inward flux of calcium linked to the EAD may cause a small but rapid increase in intracellular calcium, which may be sufficient to trigger the calcium-induced release of calcium from the sarcoplasmic reticulum, causing a much greater increase in cytosolic calcium and the occurrence of a second contraction or the prolongation of the contraction itself.

Cardiac Events and Their Relationship to Genotype

The syncopal episodes are caused by TdP often degenerating into ventricular fibrillation. Long pauses facilitate the onset of TdP.²¹ Although most LQTS patients develop their symptoms under stress, sometimes these life-threatening cardiac events occur at rest. The reason(s) for these different patterns remained obscure until molecular biology was able to distinguish among different genotypes. In 1995, an unexpected observation was made that among a tiny group of genotyped individuals, LQT2 patients appeared to be at higher risk during emotional stress, whereas LQT3 patients had their events mostly at rest or during sleep; this fostered a large and targeted study that shed light on this issue of major clinical relevance.^22,²³

In 670 LQTS patients of known genotype and who all had symptoms (syncope, CA, or sudden death), Schwartz et al examined possible relationships between genotype and the conditions (“triggers”) associated with the events.²³ As predicted by their impairment on the I_Ks current (essential for QT shortening during increases in heart rate), most of the events in LQT1 patients occurred during exercise or stress. Swimming is especially dangerous for them, and 99% of the events which occurred while swimming did involve LQT1 patients. Conversely, most of the events (including the lethal ones) in LQT2 patients occurred during emotional stress such as auditory stimuli (e.g., sudden noises and telephone ringing, especially occurring while at rest), and most of the events of LQT3 patients occurred while they were asleep or at rest (Figure 62-4).

FIGURE 62-4 Lethal cardiac events according to triggers and genotype. Numbers in parentheses are triggers, not patients.

(Modified from Schwartz PJ, Priori SG, Spazzolini C, et al: Genotype-phenotype correlation in the long QT syndrome. Gene-specific triggers for life-threatening arrhythmias, Circulation 103:89–95, 2001.)

Patients with LQT2 and LQT3 are at low risk during exercise because they have a well-preserved I_Ks current and are therefore able to shorten their Q-T interval whenever the heart rate increases. The practical implication is that young individuals with LQT2 and LQT3 should be allowed to play (no competitive sports, however) rather freely, which would give them significant psychological benefits.

In women, even during the postpartum period, genotype is important because the risk is higher for patients with LQT2 than for those with LQT1.^24,²⁵ The higher risk for women with LQT2 is probably partly related to sleep disruption; accordingly, it is recommended that the spouses or other caregivers take on some of the night-time feeding of the infants, thus allowing a fair amount of uninterrupted sleep for the women with LQT2.

These significant genotype-phenotype correlations have an impact on management.

Molecular Diagnosis in Long QT Syndrome

Who Should Be Screened for Long QT Syndrome and Medico-legal Implications

Molecular diagnosis should always be attempted whenever LQTS has either been diagnosed or is suspected on sound clinical grounds. When successful (80% to 85% of cases in the author’s laboratory), it allows the rapid screening of all the family members and the identification of all the “silent mutation carriers” (individuals with a disease-causing mutation and a QTc less than 440-ms interval) whose prevalence increases from 10% for LQT3 to 37% for LQT1 patients.²⁶

This has major clinical and medico-legal implications. For example, suppose that in a family with a boy and a girl, the boy and one parent are clearly affected (syncope and prolonged QTc). If the girl has a normal Q-T interval and is incorrectly assumed to be unaffected, she might later be treated with one of the many drugs that block the I_Kr current; she could develop TdP and die. This would not be bad luck; it would be the direct responsibility of the physician who had not had molecular screening done for the boy who was clearly affected: Identification of the disease-causing mutation would have allowed rapid determination of whether the girl was a mutation carrier, which would have saved her life. Physicians who do not attempt to genotype their LQTS probands are deliberately choosing to ignore the possibility that other family members might be mutation carriers at risk for sudden death.

Molecular Genetics and Risk Stratification

Molecular genetics contribute importantly to risk stratification. In 2003, data on 647 patients of known genotype from 193 families indicated that the incidence of life-threatening events was lower among LQT1 patients, but this was partly because of the high prevalence of silent mutation carriers (QTc < 440 ms); the risk was higher among females with LQT2 compared with males with LQT2 and among males with LQT3 compared with females with LQT3. Independent of genotype, the risk of becoming symptomatic was strongly correlated with QTc and was markedly greater with QTc greater than 500 ms. Most of the patients with LQT1 go through life without ever having cardiac events; also, among patients with LQT2 and LQT3, almost half of them remain asymptomatic. The fact that this is often forgotten is shown by the growing preference for implantable cardioverter-defibrillator (ICD) implantations in asymptomatic individuals just because they have been diagnosed with LQTS.²⁷

In 2002, Moss et al indicated that patients with LQT2 with mutations in the pore region were at higher risk compared with patients with LQT2 with mutations in different regions of the same gene.²⁸ In 2007, Moss et al demonstrated in 600 patients with LQT1 that both the transmembrane location of the mutations and their dominant-negative effect are independent risk factors for cardiac events.²⁹ Shortly thereafter, Crotti et al focused on the hot spot KCNQ1-A341V, a common mutation found responsible for a founder effect in 25 South African families, and demonstrated that the unusually high clinical severity already reported by Brink et al in the South African families is present also among patients with LQT1 from different ethnic backgrounds but carrying the same A341V mutation.^30,³¹ Moreover, as KCNQ1-A341V has a mild dominant-negative effect (the current loss barely exceeds 50%), its striking clinically severe phenotype is explained neither by the location (transmembrane) nor by the functional consequence of the mutation (dominant-negative). This implies that the current biophysical assessments of the electrophysiological effects of LQTS-causing mutations do not provide all the information necessary to make a complete genotype-phenotype correlation. In this regard, the study by Crotti et al paves the way for a mutation-specific risk stratification.³⁰

The risk stratification process is complicated by the presence of additional genetic variants that may modify clinical severity, as demonstrated in a family with LQT2 with C-terminal A1116V mutation, where the risk for life-threatening events was increased by the presence of the very common KCNH2-K897T polymorphism. Electrophysiological evidence did show that K897T produces an accentuation of the mutation-dependent I_Kr current loss resulting in the unmasking of a clinically latent C-terminal LQT2 mutation.¹³ This finding has very recently confirmed by Antzelevitch’s group in sudden infant deaths.³²

Genetic variations in NOS1AP, which encodes a nitric oxide synthase adaptor protein, contribute to Q-T interval duration in the general population.³³ Accordingly, the author’s group tested in their enlarged South African founder population (500 subjects, 205 mutation carriers) the hypothesis that NOS1AP is a genetic modifier of LQTS. The main findings were that two NOS1AP variants (rs4657139 and rs16847548) were significantly associated with occurrence of symptoms, with clinical severity manifested by an almost double probability for CA and sudden death, and with a greater likelihood of having a Q-T interval in the top 40% of the values among all mutation carriers. This is the first evidence, demonstrated in subjects sharing the same mutation, that NOS1AP is a genetic modifier of LQTS and that some of its variants are associated with a greater risk for CA and sudden death.¹⁵ One of the best evidences for the value of molecular biology in risk stratification comes from J-LN syndrome, in which a striking difference in prognosis is seen, according to the gene involved, KCNQ1 or KCNE1 (Figure 62-5).³

FIGURE 62-5 Kaplan-Meier curve of event-free survival in patients with Jervell and Lange-Nielsen syndrome, with mutations in KCNQ1 or KCNE1 genes.

(From Schwartz PJ, Spazzolini C, Crotti L, et al: The Jervell and Lange-Nielsen syndrome. Natural history, molecular basis, and clinical outcome, Circulation 113:783–790, 2006.)

Clinical Diagnosis

Given the characteristic features of LQTS, the typical cases present no diagnostic difficulty for the physicians who are aware of the disease. However, borderline cases are more complex and require the evaluation of multiple variables besides clinical history and ECG studies. Diagnostic criteria were proposed in 1985 and subsequently updated in 1993 and in 2006.^4,³⁴^–³⁶

The new diagnostic criteria are listed in Table 62-1. In the experience of the author of this chapter, the point score was arbitrarily divided into three probability categories: (1) ≤1 point = low probability of LQTS; (2) >1 to 3 points = intermediate probability of LQTS; and (3) ≥3.5 points = high probability of LQTS. Whenever a patient receives a score of 2 to 3 points, serial ECGs and especially several 24-hour Holter recordings should be obtained because the QTc value in patients with LQTS may vary from time to time and from day to night. In this group, with intermediate probability of LQTS, the presence of additional morphologic abnormalities may help in diagnostic decisions. Often, precordial leads are more informative.

Table 62-1 1993–2006 Long QT Syndrome (LQTS) Diagnostic Criteria

	POINTS
ELECTROCARDIOGRAPHIC FINDINGS*
>480 ms	3
A. QTc†
460–470 ms	2
450–459 (male) ms	1
B. Torsade de pointes‡	2
C. T-wave alternans	1
D. Notched T wave in three leads	1
E. Low heart rate for age§	0.5
CLINICAL HISTORY
With stress	2
A. Syncope‡
Without stress	2
B. Congenital deafness	0.5
FAMILY HISTORY¶
A. Family members with definite LQTS	1
B. Unexplained sudden cardiac death before age 30 years among immediate family members	0.5

Score: ≤1 point = low probability of LQTS; >1 to 3 points = intermediate probability of LQTS; ≥3.5 points = high probability of LQTS.

* In the absence of medications or disorders known to affect these electrocardiographic features.

† QTc calculated by Bazett’s formula, where .

‡ Mutually exclusive.

§ Resting heart rate below the second percentile for age.

¶ The same family member cannot be counted in A and B.

The diagnostic criteria discussed above were conceived in the premolecular era and should be used with common sense. Obviously, they cannot be of value in identifying the so-called silent mutation carriers, who have a normal Q-T interval. For these individuals, molecular screening is essential. The main value of these clinical criteria is at the time of the first contact with a patient and in clinical studies when uniformity in diagnosis is essential.

Therapy

Antiadrenergic Interventions

The most significant information on therapy still comes from a 1985 study, which included 233 symptomatic patients with detailed clinical information on the time of first syncope and with an adequate follow-up.³⁷ The mortality at 15 years after the first syncope was 9% in the group treated by antiadrenergic therapy (β-blockers, sympathectomy, or both) and more than 53% in the group not treated or treated by miscellaneous therapies not including β-blockers. These data conclusively demonstrated that pharmacologic antiadrenergic therapy, surgical antiadrenergic therapy, or both radically modify the prognosis for symptomatic patients with LQTS. In 2009, mortality among patients treated with β-blockers and left cardiac sympathetic denervation (LCSD) dropped to around 1%. The dramatic success of carefully executed therapies underscores the unacceptability of symptomatic patients with LQTS being either undiagnosed or incorrectly treated.

β-Blockers

β-Adrenergic blocking agents represent the first-choice therapy in symptomatic patients with LQTS, unless specific contraindications are present.

Propranolol and nadolol are the two most effective drugs. Their dosages are 3 mg/kg (sometimes increased to 4 mg/kg) and 1 mg/kg, respectively. The main advantages of propranolol are its lipophilicity, which allows it to cross the blood-brain barrier, and its well-known tolerability for chronic therapy; its main disadvantages are the need of multiple daily administrations and its contraindications for patients with asthma and diabetes. Nadolol is used more and more often, as its longer half-life allows twice-daily administration at a daily dose slightly lower than that of propranolol. This increases compliance, particularly in teenagers who may easily forget to take their medication in the afternoon. Unfortunately, β-blockers are not identical, and many of them are unquestionably less effective; this group includes bisoprolol, metoprolol, atenolol, and carvedilol.

In a large number of patients of unknown genotype, mortality on β-blocker therapy was 2%, and it was 1.6% when limited to patients with syncope (no cardiac arrest) and without events in the first year of life.³⁸ Clear evidence that β-blockers are extremely effective in LQT1 patients does exist. Data from two large studies reported mortality around 0.5% and sudden death combined with CA up to 1%.^23,³⁹ The impairment in the I_Ks current makes these patients particularly sensitive to catecholamines and quite responsive to β-blockade. These patients seldom need more than antiadrenergic therapy.

Particularly, important information has come from a study published in 2009.⁴⁰ Vincent et al performed a retrospective study of the details surrounding cardiac events in 216 patients with genotyped LQT1, who were treated with β-blockers and were followed up for 10 years. Before β-blocker therapy, cardiac events occurred in 157 patients (73%) at a median age of 9 years, with CA in 26 (12%). After β-blockers, 75% were asymptomatic, and the risk for life-threatening cardiac events was reduced by 97% (P < .001). Twelve patients (5.5%) had a CA or sudden death, but 11 (92%) of 12 were noncompliant (n = 8), were on a QT-prolonging drug (n = 2), or both (n = 1) at the time of the event. The risk for CA or sudden death in compliant patients not taking QT-prolonging drugs was dramatically less compared with noncompliant patients on QT-prolonging drugs (odds ratio, 0.03; 95% CI, 0.003 to 0.22; P < .001). None of the 26 patients with CA before β-blocker had CA or sudden death on β-blockers. The conclusion was that β-blockers are extremely effective in treating LQT1. Noncompliance with β-blocker therapy and use of QT-prolonging drugs are responsible for almost all life-threatening β-blocker “failures.”

Partly because of the small numbers available (which affect percentages), a concept unsupported by firm evidence has rapidly spread—that LQT3 patients are not protected at all by β-blockers or by antiadrenergic therapy. In fact, actual data point to a profound difference seen among these patients on the basis of the age of the first manifestation. On the one hand, if cardiac events occur during the first year of life, then β-blockers are insufficient and the disease has a highly malignant course. On the other hand, with a mean follow-up of 9 years, patients without cardiac events in the first year of life have done extremely well with β-blockers, LCSD, or both.⁴¹ Larger studies will have to be performed, but available evidence indicates that a molecular diagnosis of LQT3 in an asymptomatic patient should never lead to an automatic decision for ICD implantation.

Left Cardiac Sympathetic Denervation

A thorough description of LCSD has recently been published.⁴² Following a small incision in the left subclavicular region, LCSD is performed using an extrapleural approach, which makes thoracotomy unnecessary. The average time for surgery is 35 to 40 minutes. LCSD requires the removal of the first four thoracic ganglia. The cephalic portion of the left stellate ganglion is left intact to avoid Horner’s syndrome. In almost 30% of patients, a very modest (1 to 2 mm) ptosis, which can be noted only by close examination but fully escapes notice in normal social interactions, results.

The latest data published in 2004 include 147 patients with LQTS who underwent LCSD during the past 35 years.⁴³ They represented a very high-risk group, as 99% were symptomatic, their mean QTc being very long (563 ± 65 ms), 48% had a CA, and especially 75% continued to have syncope despite full-dose β-blockers. The data most relevant to current clinical decisions are those regarding patients without cardiac arrest (who almost always should receive an ICD) who have syncope despite being treated with a full dose of β-blockers. During a mean follow-up of 8 years, a 91% reduction in cardiac events was observed. LCSD produced a mean QTc shortening of 39 ms, which indicates an action on the substrate as well as on the trigger. Mortality was 3% in this high-risk group. A postsurgery QTc less than 500 ms predicted a highly favorable outcome. Importantly, this series included five patients who underwent LCSD because they had experienced multiple ICD shocks and electrical storms: in this group, over a 4-year follow-up, a 95% decrease in the number of shocks (from an average of 29 shocks per year), with a dramatic improvement in the quality of life of the patients and of their families, was observed.

The major antifibrillatory efficacy of LCSD has been previously demonstrated in high-risk patients with post–myocardial infarction; and recently, it was demonstrated in patients with catecholaminergic polymorphic ventricular tachycardia (CPVT), who were not protected by β-blockers and were receiving multiple appropriate shocks from the ICD.^44,⁴⁵ The Boston Children Hospital and the Mayo Clinic have recently published their highly successful experience with LCSD in both LQTS and CPVT.⁴⁶^–⁴⁸

Whenever syncopal episodes recur despite full-dose β-blocking therapy, LCSD should be considered and implemented whenever possible. It is unfortunate that despite its relative simplicity, LCSD is not performed in many high-risk patients because cardiovascular surgeons are no longer familiar with the procedure. The consequence is that too often the choice goes to the easiest approach, namely, ICD implantation, even when this choice is not the best for the patient. Before making a decision on therapy for a young patient who experiences syncope despite β-blockers, the patient and the family have the right to be informed about the long-term benefits and limitations of both LCSD and ICD implantation.

Implantable Cardioverter-Defibrillators

A major, largely unjustified, increase has been seen in the number of ICDs implanted in patients with LQTS. A consensus seems to exist for immediately implanting an ICD in case of a documented CA, either on or off therapy, unless the event occurs because of a transient and reversible cause. By contrast, opinions differ strongly regarding the use of ICDs in patients who have not had a cardiac arrest.

The U.S. and the European ICD-LQTS Registries provide the disquieting information that the majority of implanted patients had not had a CA and that many had not even failed β-blocker therapy.^49,⁵⁰ A recent multicenter study went so far as to indicate that the mere presence of an SCN5A mutation, even in a totally asymptomatic individual, is sufficient for immediate ICD implantation.²⁷

It should not be forgotten that ICDs do not prevent occurrence of malignant arrhythmias and that TdP is frequently self-terminating in LQTS. The recurrence of electrical storms has led to suicide attempts by teenagers; the high incidence (>10%) of electrical storms in children has been considered “devastating” by a large group of experienced pediatric cardiologists.⁵¹ The massive release of catecholamines—triggered by pain and fear that follow an ICD discharge in a conscious patient, especially a young one—leads to further arrhythmias and to further discharges, all of which produce a dramatic vicious circle.

Special caution is necessary before choosing to implant an ICD in a child with LQTS. Such a decision is seldom justified before a proper trial with combined antiadrenergic therapy, that is, full-dose β-blockade and LCSD, which prevents sudden death in 97% to 98% of symptomatic and high-risk patients and which still allow an ICD implant in case of a new syncope or CA. However, the nature of the disease is such that CAs may recur and have lethal consequences. Thus, concerns for the life of the patients and the interference of medico-legal considerations are the reality. The risk-benefit ratio of an ICD should be clearly explained to the patient or to his or her parents, and information on the pros and cons of LCSD should be provided, to allow them to make an informed choice.

The European data on 233 patients with LQTS who had an ICD indicate that 9% of these patients were asymptomatic and that 41% received an ICD without having been first on LQTS therapy.⁵⁰ Within 5 years 31% of the patients had adverse events. Appropriate ICD discharges (in 28% of patients) were predicted by age younger than 20 years at implantation, prior CA, and cardiac events despite therapy. Among patients without these factors, no one received appropriate shocks within 7 years, in contrast to 70% of those with all four factors.

The current policy of the author’s group is to implant an ICD after a CA has occurred, with the exception of reversible or transient triggers; always when firmly requested by the patient; when syncope recurs despite β-blockade and LCSD; and whenever clinical sense detects imminent danger despite therapy.

Gene-Specific Therapy and Management

Patients with LQT1 are at higher risk during sympathetic activation, such as during exercise and emotions. They should not participate in competitive sports. Swimming is particularly dangerous, as 99% of the arrhythmic episodes in patients with LQT1 were associated with swimming.²³

In patients with LQT2, some of whom have a tendency to lose potassium (K⁺), it is essential to preserve adequate K⁺ levels. Oral K⁺ supplements in combination with K⁺-sparing agents are a reasonable approach. As these patients are at higher risk, especially when aroused from sleep or rest by a sudden noise, it is recommended that telephones and alarm clocks be removed from their bedrooms; especially in the case of children, whenever they have to be wakened in the morning, it should be done gently and without yelling.²³

The realization that SCN5A mutations producing LQT3 have a “gain-of-function” effect has lent support to the early suggestion by the author’s group to try sodium channel blockers, especially mexiletine, as possible adjuvants in the management of patients with LQT3.^8,^22,³⁶ The author’s group tests the effectiveness of mexiletine in all patients with LQT3 by using the acute oral drug test technique (half the daily dose during continuous ECG monitoring). Within 90 minutes, the peak plasma concentration is reached; if the QTc is shortened by more than 40 ms, then mexiletine is added to β-blocker therapy. In fact, in most patients with LQT3, QTc is shortened by more than 70 to 80 ms with mexiletine (Figure 62-6). Even though no conclusive evidence for a beneficial effect exists and definite failures have occurred, evidence of significant benefit in a number of individual cases is growing. The highly malignant forms manifest in infancy because of mutations that cause extremely severe electrophysiological dysfunction, which is corrected by the combination of mexiletine and propranolol.⁵²

FIGURE 62-6 LQT3 patient aged 17 years, major QTc prolongation (QTc 516 ms) not modified by β-blocker therapy. The addition of oral chronic therapy with mexiletine (200 mg twice daily) produced a major shortening of QTc by over 100 ms and persisted over time. HR, Heart rate.

(From Schwartz PJ, Crotti L: Long QT and short QT syndromes. In Zipes DP, Jalife J, editors: Cardiac electrophysiology: From cell to bedside, ed 5, Philadelphia, 2009, Saunders.)

During heart rate increases, QTc shortens more in patients with LQT3 than among healthy controls; indeed, normal physical activity may not have to be restricted in patients with LQT3.²² They are at higher risk of death at rest, especially at night-time. When these patients sleep and are in a horizontal position, the onset of TdP produces a progressive but slow fall in blood pressure, which facilitates noisy gasping preceding death. Patients with LQT2 (also at risk while resting) as well as those with LQT3 have, on occasion, been saved by family members, for example, the spouse sleeping next to the patient. Therefore, it is recommended that patients with LQT2 and LQT3 keep an intercom system in their bedrooms and, if young, in the parents’ or other caregivers’ bedrooms.