Association of Long QT Syndrome Loci and Cardiac Events Among Patients Treated With β-Blockers FREE

Long QT syndrome (LQTS) is a genetic disease characterized by prolonged ventricular repolarization, syncope, ventricular arrhythmias, and sudden death,^1- 3 often precipitated by emotion or exercise. Primarily according to nonrandomized trial evidence, β-blockers are considered first-line prophylactic therapy,⁴ whereas patients refractory to β-blockers may be treated with left-sided cardiac sympathetic denervation, pacemakers, or implantable cardioverter defibrillators.^1,5- 8 The hypothesis that the efficacy of therapy may vary according to the genetic form of the disease has been proposed⁷ but not thoroughly investigated.

Three genetic loci account for nearly 98% of genetically characterized patients. In this investigation, we sought to describe and assess outcomes of β-blocker–treated patients affected by the 3 most common genetic loci of LQTS⁹: LQT1, LQT2, and LQT3, caused by genetic defects on KCNQ1, KCNH2, and SCN5A genes.

The study population included 335 genotyped LQT1, LQT2, or LQT3 patients treated with long-term β-blocker therapy. For each patient, data on personal and family history, cardiac events, and therapy were systematically recorded at each visit or medical contact. The specific β-blocker used, as well as dose, was at the discretion of the treating physician. LQTS-related cardiac events included unexplained syncope, torsades de pointes, ventricular tachycardia, aborted cardiac arrest, and sudden cardiac death. All patients or their guardians provided written informed consent for clinical and genetic evaluation. The protocol was approved by the institutional review board of the Policlinico S. Matteo and of the Maugeri Foundation, Pavia, Italy.

Patients were consecutively genotyped at the Molecular Cardiology Laboratories of the Maugeri Foundation as carriers of a single mutation on KCNQ1, KCNH2, or SCN5A genes; carriers of double mutations, representing on average 3% to 5% of the patients, were excluded from the study.

DNA was extracted from peripheral blood lymphocytes and amplified with primer pairs for KCNQ1, KCNH2, and SCN5A. Genetic analysis was performed by standard methods: denaturating high-performance liquid chromatography (Wave Transgenomics, Omaha, Neb) was performed on polymerase chain reaction–amplified DNA, encompassing the entire open reading frame of each gene, by using intronic primers. When abnormal chromatograms were identified, double-strand sequencing of amplified genomic DNA (ABI Prism 310; Applied Biosystems, Foster City, Calif) of the corresponding amplicon was performed. In addition to the previously reported polymorphisms, all DNA variants causing coding variations and occurring in more than 1 in 100 of the control population (400 healthy controls; ie, 800 alleles) were considered polymorphisms.

All analyses were performed with the SPSS 11.0 statistical package (SPSS Inc, Chicago, Ill). Statistical significance was set at P≤.05. Genetic loci–related differences for continuous variables were assessed by using 1-way analysis of variance, with post hoc analysis with the Tukey test, whereas differences for categorical variables were assessed with the Pearson χ² test. Survival analyses included construction of Kaplan-Meier plots with comparisons with the log-rank χ² test, as well as forward-selection Cox proportional hazards modeling. We considered sex, QT interval corrected for heart rate (QTc), occurrence of cardiac events before therapy, age at first cardiac event before therapy, family history of sudden death, and genotype as candidate variables.

The 335 genotyped LQTS patients were from 187 families with mutations on KCNQ1 (LQT1; n = 187), KCNH2 (LQT2; n = 120), or SCN5A (LQT3; n = 28) genes treated with β-blockers. Before therapy, 159 of 335 (47%) experienced cardiac events. The mean (SD) age at initiation of β-blocker therapy was 21 (17) years (interquartile range, 8.5-31.7 years); the median follow-up for patients without events and receiving β-blocker therapy was 4.7 years (range, 0.6-36 years). As summarized in Table 1, no differences among LQT1, LQT2, and LQT3 patients were observed in age, age at initiation of therapy, observation time while receiving β-blockers, and age at first cardiac event before therapy. However, LQT1 patients had a shorter QTc interval. Data for type of β-blocker and dosage per kilogram of body weight were available for 266 individuals: 69% of them were treated with either propranolol (average daily dose, 2.2 [SD, 1.04] mg/kg) or nadolol (average daily dose, 1.2 [SD, 0.5] mg/kg ); there were no dosage differences among the 3 genotypes (P = .31).

There were 55 patients (16%) who experienced cardiac events while receiving β-blocker therapy, of whom 14 (25%) had a cardiac arrest; 4 sudden cardiac deaths occurred (1 LQT1 and 3 LQT3). Events were not evenly distributed in the 3 loci, with LQT1 having the lowest incidence of cardiac events (LQT1: 19/187 [10%]; LQT2: 27/120 [23%]; and LQT3: 9/28 [32%]; P = .001) (Table 1 and Figure 1).

In a multivariable model, important predictors of time free of events were first cardiac event before therapy in early childhood (age ≤7 years), QTc more than 500 ms in patients receiving therapy, and genetic locus other than LQT1 (Table 2). When we combined patients with LQT2 and LQT3, they were at substantially increased risk for cardiac events compared with those with LQT1 (18/187 [10%] vs 36/148 [24%]; odds ratio, 2.8; 95% confidence interval [CI], 1.6-5.2; P<.001). Similarly, LQT2 and LQT3 patients were at increased risk for cardiac arrest or sudden cardiac death compared with those with LQT1 (8% vs 1%; odds ratio, 8.1; 95% CI, 1.8-37; P = .001). Four of 14 patients experiencing cardiac arrest while receiving therapy had been asymptomatic before therapy; thus, cardiac arrest was their first symptom of LQTS.

These results were confirmed in a supplementary analysis of the 187 probands of each family to correct for a possible confounding effect introduced by the presence of family members. In another supplementary analysis, we found that LQT2 or LQT3 genotype was associated with a higher risk of cardiac events in patients receiving β-blockers, even after excluding patients who had a cardiac arrest before therapy (49 events: 2/183 LQT1, 7/110 LQT2, and 3/23 LQT3; odds ratio, 2.15; 95% CI, 1.2-4.2; P = .002).

Prophylactic therapy with β-blockers is the mainstay treatment for LQTS patients. In our large series, however, we found a high rate of cardiac events in patients receiving β-blocker therapy, particularly for patients with LQT2 and LQT3 genotypes. These observations are troubling and suggest that, for some patients, genotyping may be useful for identifying candidates for more aggressive interventions, possibly including defibrillator implantation.

Our findings are consistent with existing evidence that genetic background may influence the severity of the disease and its clinical manifestations before treatment.¹⁰ In 1985, Schwartz⁴ provided the first data indicating that a subset of LQTS patients was not fully protected by β-blockers. In 2000, Moss et al⁷ further quantified this worrisome phenomenon and, reporting data on 139 genotyped patients, introduced the concept that the response to β-blockers may be modulated by genetic substrate. We have extended those observations by showing in a large systematically genotyped β-blocker–treated cohort a gradient of risk from LQT1 to LQT2 to LQT3 genotypes. Furthermore, we have shown that in addition to genotype, other important predictors of cardiac events in patients receiving therapy include QTc interval and younger age at a first pretherapy cardiac event.

Because no randomized trial data exist, our findings about the value of defibrillator implantation in β-blocker–treated LQTS patients who have not experienced a cardiac arrest cannot be considered conclusive. Nonetheless, these findings suggest that prophylactic defibrillator therapy may be a reasonable addition to β-blockers in patients with LQT2 or LQT3 genotypes. However, a decision to implant a defibrillator in LQT2 and LQT3 patients in the absence of definitive randomized trial evidence is a complex one that requires consideration of important issues such as the implantable cardioverter defibrillator's impact on quality of life of young patients, high complication rates, the acceptance of the device by patients and their families as an alternative, and the incomplete but additional protection by left-sided cardiac sympathetic stimulation.⁵ Furthermore, it is unlikely that β-blocker therapy could be safely discontinued after defibrillator implantation.

This study is based on an observational registry and is subject to all the inherent limitations of such an analysis. Because LQTS is a relatively rare disease, it is unlikely that large-scale randomized trial data will become available soon, meaning that evaluation and treatment of these patients must occur in a setting of incomplete evidence. Furthermore, approximately 40% of LQTS patients cannot be genotyped on the known loci, so for these patients, β-blockers remain the recommended therapy.