Congenital long QT syndrome (LQTS) is an inherited arrhythmic disorder clinically characterized by delayed repolarization of cardiomyocytes, resulting in prolongation of the QT interval and a predisposing to life-threatening ventricular arrhythmias and sudden cardiac death (SCD) (1, 2). Pathogenic DNA variants in genes encoding ion channels involved in cardiomyocyte action potential formation are the primary culprits in the development of LQTS. The most commonly affected genes are the so-called significant LQTS genes KCNQ1, KCNH2, and SCN5A (3). Historically, the prevalence of the disease was estimated at 1:20,000 to 1:5,000, but subsequent data from Schwartz et al. suggested a prevalence of 1:2,000 (4).

LQTS remains a significant cause of sudden cardiac death in young individuals. Early diagnosis is essential to prevent malignant arrhythmias, enable accurate risk stratification, and identify relatives who may also carry pathogenic variants. Therapy with beta-blocking agents has been proven to significantly reduce the risk of life-threatening ventricular arrhythmias (5).

Over recent decades, several cohorts and registry studies have substantially advanced our understanding of the clinical risk stratification of congenital LQTS, revealing that therapeutic responses to different beta-blockers (BBs) vary across patients. Consistently identified predictors of major arrhythmic events include pronounced QTc prolongation, elevated Schwartz score, symptomatic presentation with syncope or aborted cardiac arrest, onset in infancy or early childhood, and LQT3 associated with SCN5A variants (1, 3, 6, 7).

More recent comparative work has demonstrated that BB therapy does not offer uniform protection in LQTS. Several multicenter analyses have reported a greater likelihood of breakthrough events among patients treated with cardioselective agents, particularly metoprolol, compared with older non-selective beta-blockers such as propranolol and nadolol. Propranolol has also been associated with more pronounced QTc shortening in higher-risk subgroups (6, 7). Meta-analyses and additional cohort data suggest that nadolol may be especially effective in LQT1 and LQT2, reinforcing the notion that BB efficacy is, at least in part, genotype-dependent. These observations have led many clinicians to preferentially prescribe non-selective BBs for pediatric and genotype-positive patients in recent years.

Despite these insights, direct comparisons of BB subtypes remain limited. Most prior investigations have focused on first arrhythmic events or composite outcomes, rather than evaluating the cumulative arrhythmic burden across different treatment phases. Our study addresses these gaps by examining both the time to first major arrhythmic event (MAE) and the overall MAE burden during periods of selective vs. non-selective BB therapy in a well-defined cohort of childhood-diagnosed LQTS.

The purpose of this study was therefore to describe the long-term clinical course of pediatric-onset LQTS, identify major risk factors for life-threatening arrhythmic events, and compare the effectiveness of selective vs. non-selective BB therapy.

Over the past few decades, considerable progress has been made regarding the diagnosis and management of LQTS. With appropriate treatment, an early and precise diagnosis of this condition can effectively prevent malignant cardiac events in these patients.

The first studies suggested a very malignant natural outcome in untreated patients, with mortality rates of 71% in LQTS patients (8). BB therapy has since been established as an effective way of preventing SCD (5, 13, 16, 18). The survival probability of the reported cohort treated with BBs ranged between 97.7% and 90.2% over 5–20 years of follow-up. Given the cohort character, these figures represent a real-world long-term survival scenario in the era of BB treatment in clinical practice. The studied group appears representative of the general LQTS population, with 43.6% of the patients symptomatic at presentation (12.9% with aborted cardiac arrest) (19, 20). Consistent with published studies (21), our cohort's most commonly affected genes were (in descending order) KCNQ1, KCNH2, and SCN5A. The yield of genetic testing (86.7%) matched the reported data in patients with a well-marked phenotype (22).

In line with available studies (3, 23, 27), risk factors for MAEs included early presentation, VT or SCA at presentation, LQTS type 3, higher Schwartz score, and longer QTc.

Previous studies have shown that patients with LQTS type 1 generally have a more favorable prognosis, with higher overall and MAE-free survival rates, compared with individuals with LQTS types 2 and 3 (3). In our cohort, however, no statistically significant differences in overall survival probability were observed across the three major genotypes (Figure 3). This finding may, in part, reflect the effectiveness of contemporary genotype-specific management strategies, but it is also likely influenced by the limited sample sizes within each LQTS subtype.

In contrast to previous studies, we evaluated the protective BB effect in terms of the first MAE episode and analyzed its efficacy in reducing the cumulative MAE burden using a Poisson regression model. Such analysis allowed us to examine various treatment periods in the same patient who switched from one BB type to another, thereby better reflecting true long-term treatment efficacy. This statistical approach confirmed a clear benefit of non-selective BBs. The only patients experiencing MAEs on non-selective BBs were those with highly malignant phenotypes presenting very early in life, with most carrying an SCN5A mutation. To the best of our knowledge, this is the first study using a Poisson regression model in LQTS patients.

Our finding that treatment with non-selective BBs was significantly more protective, regardless of other risk-determining variables, and was associated with a lower cumulative MAE burden extends prior comparative studies (6, 7). Chockalingam et al. observed a higher recurrence of cardiac events under metoprolol compared with propranolol and nadolol, with greater QTc shortening reported with propranolol. Abu-Zeitone and colleagues likewise compared different BBs within an international registry and found genotype-dependent differences in efficacy, with nadolol performing exceptionally well in the LQT1 and LQT2 subgroups. Our results are consistent with these reports and make an additional novel contribution: The effect persists when MAE burden (recurrent events over time) is modeled using a Poisson regression model, and the benefit is observed in the pediatric-diagnosed cohort followed into adolescence/adulthood.

Several pharmacodynamic properties may explain why non-selective β-blockers (e.g., nadolol, propranolol) appear more effective than cardioselective β1 agents (e.g., atenolol, metoprolol) in reducing MAEs in LQTS. Nadolol and propranolol block both β1 and β2 receptors. In specific LQTS triggers (sympathetic response, exercise), β2 blockade in the myocardium and vascular system may provide additional stabilization of repolarization and reduce heterogeneity of ventricular refractoriness beyond β1 blockade alone (24). Several selective agents or older drugs (e.g., acebutolol, pindolol) also exhibit intrinsic sympathomimetic activity (ISA), which can produce partial agonist effects and may thus be less suppressive of adrenergic-mediated arrhythmogenesis. Most non-selective agents used in reported LQTS cohorts (nadolol, propranolol) lack ISA (25). Moreover, propranolol has been shown to inhibit voltage-gated sodium channels in cardiac tissue at therapeutic concentrations, producing a direct antiarrhythmic membrane-stabilizing effect that is independent of β-receptor antagonism. This mechanism may be particularly relevant for arrhythmias arising from sodium current abnormalities (including some SCN5A phenotypes) or for suppressing triggered activity (26). In addition, nadolol has a long half-life and relatively stable plasma levels with once-daily dosing, which can provide more consistent adrenergic blockade over 24 h compared with shorter-acting agents (e.g., metoprolol), particularly if dosing adherence is imperfect (7). Together, these properties explain the superior effectiveness of non-selective BBs observed in multiple cohorts and in the present analysis.

While BBs are highly effective in LQT1 (IKs channel defects), their effectiveness is less pronounced in LQT3 (SCN5A gain-of-function). The few MAEs observed on non-selective agents in our cohort clustered in early-presenting, highly malignant SCN5A phenotypes, consistent with prior reports that SCN5A carriers remain at relatively higher residual risk despite therapy (26).

In our cohort, the Poisson model estimated ∼0.085 MAEs per patient over 10 years on non-selective BB vs. ∼0.327 MAEs per patient on selective BB (p = 0.011), a clinically meaningful difference that aligns directionally with those prior reports and with meta-analytic syntheses. Notably, since a subset of patients switched between BB types during follow-up, our within-patient modeling of treatment periods reinforces the argument that the observed difference is driven by drug characteristics rather than solely by baseline patient selection.

According to the results of the mentioned studies, many centers now use non-selective BBs as the preferred first-line treatment for high-risk LQTS patients, particularly in children and in patients with LQT1 or LQT2. If nadolol cannot be used, propranolol is a reasonable alternative, provided adequate dosing is ensured, to achieve a 24-h beta-blockade. For patients with SCN5A-related LQT3, BBs offer less protection overall, so additional genotype-specific treatments (such as mexiletine) or device therapy may be necessary (7).

BBs are the mainstay of therapy for LQTS and are recommended for nearly all affected children; however, specific scenarios require avoidance, dose adjustment, or careful monitoring. Neonates and infants with very low resting heart rates are at risk of pauses that may trigger TdP, which can be prevented with antibradycardia pacing. This was implemented in 13 patients in our cohort. No sinus node dysfunction was observed.

Exercise intolerance was managed by adjusting the dose, guided by repeated exercise testing, to maintain adequate β-blockade. Bronchospasm, a known adverse effect of non-selective BBs, was addressed through dose modification or, in some cases, switching to selective BBs. These complications were infrequent in the pediatric population.

Direct randomized comparisons of specific BB therapy remain lacking. Available data are observational and subject to confounding by indication, era (drug availability and prescribing patterns have changed over time), and dosing differences. Our retrospective design is therefore susceptible to such biases despite statistical adjustment and within-patient treatment period modeling. Nevertheless, the consistency of results across different centers, combined with our own Poisson model analysis of cumulative burden, makes a compelling case for preferential use of non-selective BBs in LQTS patients.

One of the significant limitations of the study is its retrospective character. Six-monthly to annual follow-ups were available for all patients during the pediatric period (i.e., less than 19 years of age). However, to obtain the follow-up data for the adult patients, we had to contact them via structured phone calls, where some information may have been missed. Moreover, in some patients, only scanned ECGs of poor quality were available, which may have led to inaccuracies in the QTc measurements.

The survival analysis in this group of patients reflects the course of the disease in patients with a definite or highly probable diagnosis of LQTS (carrying a likely pathogenic/pathogenic variant in one of the genes associated with LQTS or a Schwartz score ≥3.5 points) who were subjected to BB treatment in the vast majority. Other patients with milder/questionable phenotypes and a negative genotype were excluded, which could potentially interfere with the results.

Survival may have been influenced by complications associated with cardiac implantable electronic devices used in a subset of the cohort. As only data on all-cause mortality have been available without specific reasons for death, the influence of the respective factors could not be analyzed.

Patients with LQTS diagnosed during childhood and treated with BBs in the vast majority had a long-term survival probability of about 90%. In our study group, early presentation, QTc duration, higher Schwartz score, symptoms at presentation, and LQTS type 3 were major predictors of MAEs. Genetic testing demonstrated a high diagnostic yield. Over time, BB therapy has shifted to non-selective BB. Non-selective BB therapy was more effective than selective BB in significantly decreasing the MAE burden, regardless of other risk-determining variables.