Author + information
- Received February 22, 2018
- Revision received May 30, 2018
- Accepted June 7, 2018
- Published online September 17, 2018.
- Alan Sugrue, MBBCha,
- Ram K. Rohatgi, MDb,
- Martijn Bos, MD, PhDb,c,
- Vaibhav R. Vaidya, MBBSa,
- Samuel J. Asirvatham, MDa,b,
- Peter A. Noseworthy, MDa and
- Michael J. Ackerman, MD, PhDa,b,c,∗ ()
- aDepartment of Cardiovascular Medicine, Division of Heart Rhythm Services, Mayo Clinic, Rochester, Minnesota
- bDepartment of Pediatric and Adolescent Medicine, Division of Pediatric Cardiology, Mayo Clinic, Rochester, Minnesota
- cDepartment of Molecular Pharmacology & Experimental Therapeutics, Windland Smith Rice Sudden Death Genomics Laboratory, Mayo Clinic, Rochester, Minnesota
- ↵∗Address for correspondence:
Dr. Michael J. Ackerman, Genetic Heart Rhythm Clinic and the Mayo Clinic Windland Smith Rice Sudden Death Genomics Laboratory, Guggenheim 501, Mayo Clinic, 200 First Street Southwest, Rochester, Minnesota 55905.
Objectives This study sought to determine the prevalence of early repolarization pattern (ERP) within a large cohort of patients with long QT syndrome (LQTS) and examine the correlation and clinical significance of ERP with symptomatic status and subsequent risk of breakthrough cardiac events (BCEs).
Background The electrocardiographic ERP is associated with an increased risk of arrhythmic events and sudden cardiac death.
Methods ERP was defined as an end-QRS notch or slur on the downslope of a prominent R-wave with a J point ≥0.1 mV in 2 or more contiguous leads of the 12-lead electrocardiogram, excluding V1 to V3. A patient was considered previously symptomatic if they had a suspected LQTS-triggered cardiac event prior to diagnosis. BCEs were defined as LQTS-attributable syncope/seizures, aborted cardiac arrest, appropriate ventricular fibrillation–terminating implantable cardioverter-defibrillator shocks, and sudden cardiac death following diagnosis and institution of a LQTS-directed treatment program.
Results In this study, 528 patients (57% female) with genotype-confirmed LQTS (283 with LQT1, 193 with LQT2, and 52 with LQT3) were reviewed from which 2,618 electrocardiograms were analyzed over a median follow-up of 6.7 (interquartile range, 3.6 to 10 years) years. Eighty-two (15.5%; female 51%) patients were identified as having ERP; 40 (50%) of these ERP-positive patients showed persistent ERP. One hundred twenty-four patients (23.5%) were classified as previously symptomatic LQTS and 39 (7.2%) experienced a subsequent BCE. ERP was not associated with either symptomatic status (p = 0.62) or BCE (p = 0.61).
Conclusions Although ERP is common in LQTS, this extensive study suggests that the presence of concomitant ERP does not correlate with either those with a history of LQTS-triggered events prior to diagnosis or those with subsequent BCEs from their treated LQTS substrate.
Risk stratification in long QT syndrome (LQTS) is challenging because incomplete penetrance and variable expressivity drives patients on vastly different clinical courses and forms of disease severity. Therefore, identification of patients at highest or lowest risk of a sentinel cardiac event or a subsequent breakthrough cardiac event (BCE) is an essential aspect of LQTS management and overall care. Enhanced risk stratification would enable intensification of treatment to those at highest risk, whereas those at lower risk can be managed appropriately and reassured without further invasive procedures, such as left cardiac sympathetic denervation or implantable cardioverter defibrillator (ICD). Although the overall outcomes in LQTS have improved considerably over the past decade (1), ongoing residual risk results in some patients experiencing LQTS-triggered BCEs despite optimal LQTS-directed therapy. Consequently, there is a continuing need to identify additional means to further enhance risk stratification.
In patients with repolarization abnormalities, early repolarization pattern (ERP) on the surface electrocardiogram (ECG) has demonstrated an association with arrhythmic events (2–6) and has shown some promise, but conflicting results, in small studies as a potential risk modulator in LQTS (7,8). However, the relationship between the transient outward K(+) current channel Ito (electrophysiology basis for the appearance of ERP) and LQTS, to the best of our knowledge, has not been examined. Consequently it is unclear the impact or function of Ito in the presence of LQTS, and whether early studies suggesting increased risk of arrhythmic events with the presence of ERP in LQTS project a mere reflection of general population ERP risk of arrhythmic events. Considering the limitations of the prior small studies and the unclear electrophysiological interaction, the aim of our study was not only to determine the prevalence of ERP in a large well-characterized cohort of patients with type 1, type 2, and type 3 LQTS (LQT1–3) but also to examine its clinical significance in terms of association with previous symptomatic versus asymptomatic status and correlation with subsequent BCEs following diagnosis and establishment of their LQTS-directed treatment program.
This study was approved by Mayo Clinic Institutional Committee on Human Research. We performed a retrospective review of the electronic medical records of 592 patients with genetically confirmed LQTS who were evaluated and treated for LQTS at Mayo Clinic’s Genetic Heart Rhythm Clinic between the years of 1999 and 2015. From this cohort, we focused on 547 patients who were of LQTS subtypes 1–3 (LQT1 in 287 [52.5%], LQT2 in 204 [37.3%], and LQT3 in 56 [10.2%]) with other subtypes excluded due to low numbers. From these 547 patients, 528 had interpretable ECGs. The electronic medical record was reviewed for demographics, clinical symptomatology, family history, genetic studies, LQTS-directed therapy, and occurrence of LQTS-related BCEs. A patient was considered symptomatic if he or she had a LQTS-attributable cardiac symptom prior to diagnosis (fetal arrhythmia, arrhythmogenic syncope or seizure, or cardiac arrest). Patients with BCEs were defined as those patients who experienced a subsequent LQTS-triggered syncope, seizure, documented ventricular tachycardia (VT), appropriate VT/ventricular fibrillation (VF)–terminating ICD shocks, aborted cardiac arrest, and/or sudden cardiac death following diagnosis and implementation of their LQTS-directed treatment program. Electrocardiographically concealed LQTS patients were defined as those with a corrected QT interval (QTc) <460 ms for women, <450 ms for men, or <440 ms for either sex ≤12 years of age on their resting 12-lead ECG.
Early repolarization definition
Considerable variation in the definition of ERP led to the publication of two recent consensus articles that aimed to provide crucial clarification (9,10). Based on these consensus documents, we defined ERP as present if all of the following criteria were met: 1) presence of an end-QRS notch or slur lying entirely above the baseline, on the downslope of a prominent R-wave (Figure 1); 2) J peak (peak of notch or slur) is ≥0.1 mV in 2 or more contiguous leads of the 12-lead ECG, excluding leads V1 to V3; and 3) QRS duration is <120 ms.
If the ERP definition was fulfilled, ERP was then further classified by lead location; inferior: leads (II, III, aVF), lateral leads (I, aVL, V5, V6), or both, and whether notching or a slur was present. Further measurements of the amplitude of the J point (Figure 2) were also undertaken by the first author (A.S.) using digital calipers built into the MUSE ECG system (GE healthcare, Chicago, Illinois). The patient’s ECG with the most manifest ERP was used for analysis.
All available 12-lead ECGs were analyzed, with the reviewers blinded to demographic, clinical, and genetic data as well as the presence of symptoms or BCE. The patient’s QTc was derived from their first Mayo Clinic ECG. The QTc was calculated (Bazett’s formula; QTc=QT√RR) using the clinical 12-lead ECG (Marquette 12SL ECG Analysis Program, General Electric Healthcare, Chicago, Illinois), which measures the QT interval from the earliest detection of depolarization in any lead to the latest detection of repolarization in any lead. ERP was identified using the above-mentioned definition. Interobserver variation was calculated between 2 observers (A.S. and V.V.) using a random sample of 100 ECGs with a calculated kappa coefficient of 0.886, with no specific ERP subtype as a cause for disagreement.
Baseline variables are presented as a number and percentage or with median and quartiles, as appropriate. To analyze differences between patients with ERP and those without ERP, a Student t test for unpaired data was performed. If required, a Wilcoxon test was used for nonparametric measures. Categorical variables were analyzed using the chi-square test or the Fisher exact test, when necessary (cell frequency of <5). Acknowledging the difficulty and potential for subjectivity when determining the presence of true arrhythmogenic syncope, we performed an additional analysis excluding syncope as an event, which showed no impact on either symptomatic status or BCE (p = 0.45 and p = 0.85, respectively). JMP Pro 10 (JMP Pro, Version 10. SAS Institute Inc., Cary, North Carolina) was used for all analyses and p < 0.05 was considered to be statistically significant.
As presented in Table 1, 2618 ECGs were analyzed from 528 patients (mean, 6 ECGs per patient; range, 1 to 21) with genotype-confirmed LQTS. In this study, 57% (n = 305) of the patients with LQTS were female with a mean age of 21 ± 16 years. The median follow-up was 6.7 (interquartile range [IQR], 3.6 to 10) years with no difference in follow-up between those with ERP versus no ERP (7 [IQR, 3.25 to 10] years vs. 6 [IQR, 3.4 to 9.6] years, respectively). Most patients were LQT1 (n = 283; 54%), followed by LQT2 (n = 193; 37%) and LQT3 (n = 52; 9%). One hundred twenty-four patients (23.5%) were classified as previously symptomatic with most patients experiencing syncope/seizure (80%), followed by fetal bradycardia (11%) and cardiac arrest (9%). Thirty-nine patients (7.2%) experienced at least 1 BCE since their first Mayo Clinic evaluation. Of those patients who experienced a BCE, 18 (46%) experienced 1 event, 18 (46%) experienced between 2 and 10 events, and 3 (8%) experienced >10 events. Most patients experienced a VF-terminating ICD shock (n = 25; 64%) following by arrhythmogenic syncope/seizure (n = 12; 30%) and aborted cardiac arrest (n = 2; 5%).
Prevalence and phenotype of early repolarization in LQTS
Eighty-two patients (15.5%; female 51%) satisfied the electrocardiographic definition of ERP and 40 (50%) of these patients showed persistent ERP across multiple ECGs. Thirty-nine patients (47.6%) had terminal QRS slurring, 36 notching (43.9%), and 7 patients (8.5%) exhibited both. The mean J point amplitude was 1.77 mm (range, 1 to 3.3 mm). ERP manifested in the inferior leads in 37/82 patients (45%), followed by the lateral leads (n = 31; 37.8%) and inferolateral leads (n = 14; 17%). ST segment was mostly ascending (n = 49; 60%) compared with horizontal or downsloping (n = 33; 40%).
Clinical significance of ERP in patients with LQTS
Compared with the ERP-negative LQTS patients, there were no differences in sex, age, LQT subtype, beta-blocker usage, QTc, or heart rate among the subset with ERP-positive LQTS (Table 1). In addition, ERP was not associated with either symptomatic status (p = 0.62) or BCE (p = 0.61; Figure 3). Additionally, the location, morphological type of ERP, or ST segment slope had no impact upon either symptomatic status (p = 0.13, p = 0.06, and p = 0.71, respectively; Figure 4) or BCE (p = 0.36, p = 0.58, and p = 0.92, respectively; Figure 5).
Subset analysis of previously symptomatic patients with LQTS and patients with subsequent BCEs
Symptomatic patients who had ERP (n = 21; 17%) showed no differences in baseline demographics compared with those without ERP (Table 2). Furthermore, patients who experienced at least 1 BCE who had ERP (n = 5; 13%) showed no differences in baseline demographics compared with those without ERP (Table 3). When examined by the type of symptom that LQTS patients experienced, there was no relationship detected for cardiac arrest (p = 0.39), syncope/seizure (p = 0.58), or fetal bradycardia (p = 0.33) among the ERP-positive LQTS patients.
J point elevation >2 mm
In those who exhibited J point elevation of ≥2 mm, we observed no difference in baseline demographics, symptomatic status or BCEs compared with those ERP-positive patients whose J point elevation was 1 to 1.9 mm (Online Table 1).
Electrocardiographically concealed LQTS and ERP
One hundred thirty-seven patients (25.9%) were defined as having an electrocardiographically concealed phenotype, with 19 (14%) being ERP-positive. Among this subset, there was no relationship observed between ERP and symptomatic status or BCE (p = 0.58 and p = 0.53, respectively) (Online Table 2).
Seventy LQTS patients (85%) with ERP had more than 1 ECG with 51 of 70 (72%) patients having ERP detected on their first ECG and 30 of 70 (42%) showing persistent ERP across all ECGs. Of those who did not have ERP detected on their first ECG (n = 31), the ERP changes were not persistent across multiple ECGs, whereas of those who had ERP on the first ECG (n = 39), 30 (76%) showed persistence of ERP across all ECGs. Nevertheless, those patients with dynamic ERP changes had no detectable differences compared with those who had ERP persistence, regarding age (p = 0.10), sex (p = 0.99), LQT type (p = 0.55), symptomatic status (p = 0.19), or BCE (p = 0.12).
This study demonstrates that ERP is a common finding in patients with LQTS with at least 1 of every 6 patients with LQTS exhibiting either transient or persistent ERP. However, the presence of ERP in patients with LQTS does not indicate a poor prognosis in terms of association with being either asymptomatic or previously symptomatic or experiencing a BCE while being treated. Although previous smaller studies have suggested that ERP is correlated with symptomatic status (7,8), this more extensive study disputes this and additionally highlights that ERP does not correlate with BCE.
It is noteworthy that our study reports a high prevalence of ERP among patients with 1 of the 3 most common LQTS genotypes. Our reported prevalence of 15.5% is higher than reported in the general population but similar to other LQT studies (7,8). The reported prevalence of ERP in the general population tends to vary by the ERP definition applied. Early studies that focused purely on ST-segment elevation alone reported a prevalence of 25% to 35% (11,12), whereas with current more focused definitions the estimated prevalence is generally between 0.9% and 5% (13–16). Why patients with LQTS have a seemingly higher prevalence of ERP than current prevalence estimates of ERP is unclear. We believe that it is likely due to the age of patients with LQTS because it is well described that ERP is more prevalent in younger patients (15,17) with a recent study demonstrating an ERP prevalence of 40% in otherwise healthy children (18). Additionally, although we did not specifically observe any relationship between heart rate and ERP in our study, another consideration is that it could be potentially related to heart rate. The majority of these patients with LQTS were taking beta blockers, and it is well known that ERP manifestation is accentuated with bradycardia (13,16).
Contrary to this large study, 2 studies previously suggested a potential role for ERP in the risk stratification of patients with LQTS. Laksman et al. (7) published an initial retrospective study on the prevalence of ERP and examined its association with symptomatic status (defined as cardiac syncope, documented polymorphic VT, or resuscitated cardiac arrest). In 113 patients (compared with 528 patients in our present study), they reported a relatively large prevalence of ERP (50 patients; 44%) of which 15 patients were classified as symptomatic. They noted no difference between symptomatic and asymptomatic status concerning the presence of ERP but when ERP was classified as minor (>1 but <2 mm) and major (≥2 mm) they noted significantly more patients with major ERP were symptomatic (58% vs. 20%; p = 0.001). Although our results validate their initial findings that minor amplitude ERP (>1 but <2 mm) is not associated with symptomatic status, we demonstrate, in a larger cohort, that even major amplitude ERP (≥2 mm) is also not associated with symptomatic status in this LQTS cohort.
The latest study by Hasegawa et al. (8) also examined the association of ERP with arrhythmic events in a population of 264 patients with LQTS. In their study, the prevalence of ERP was 21%, and 135 (51%) patients were classified as symptomatic. Multivariate analysis showed that ERP was independently associated with arrhythmia events. It is notable that more patients in this study had arrhythmic events (n = 135; 51%) than did not (n = 129), which is a surprisingly large number given that the baseline symptomatic status was similar between our article and the article by Laksman et al. (7), 23% and 22%, respectively. This high event rate could indicate a higher-risk LQTS population, incomplete medical therapy, referral bias, or the fact that a different definition or ascertainment of events was used. Therefore, based on our large well-characterized cohort, we feel that the presence of ERP in LQTS should not alter management strategies either with intensification of treatment or consideration of deintensification.
Although our results suggest no role for concomitant ERP in risk stratification compared with other published studies, it is also vital to be cognizant that ERP is a dynamic ECG feature. Not only is it dynamic in its occurrence but also its location and extent. Some studies have provided valuable insight into these changes over time, but it is not something consistently examined or reported across all studies. One of the first studies to address the dynamics of ERP was by Tikkanen et al. (16) who showed that in 542 patients with J-point elevation on the baseline ECG, ERP pattern was again observed in 443 patients (81.7%) who underwent a repeat ECG examination (an average of 5 years after baseline). Noseworthy et al. (19) further showed that exercise training led to significant increases in the prevalence of ERP among athletes. Additionally, the CARDIA (Coronary Artery Risk Development in Young Adults) study showed that the prevalence of ERP diminished over time, from 24.8% to 14.7% at 7 years from initial ECG and then 6.6% at 20 years (20). Our data is consistent in this regard because only 40 (50%) of our patients showed persistent ERP over time across multiple ECGs. This dynamic nature of ERP creates implicit problems that may hamper ERP’s potential value as a risk prediction tool. When appraising studies that highlight that ERP may be associated with a risk of VF and the authors rely on the location of early repolarization, the magnitude of the ERP, and the degree of ST elevation to define these risks, particular consideration needs to be paid to the dynamics of this ECG feature.
The electrophysiological basis for the appearance of ERP on the surface ECG is a change in repolarization with relative accentuation of the transient outward K(+) current channel (Ito) activity. The increased net repolarization current results in an augmented voltage gradient causing an accentuation of the phase 1 notch of the action potential. The relationship between Ito and LQTS, to the best of our knowledge, has not been studied. It is unclear the impact or function of Ito in the presence of LQTS (particularly the delayed rectifier K+currents, IKr and IKs). From our clinical data, it would seem that there is limited channel interaction or implications on function. Further functional studies are required to examine this.
Although this study is the largest to date regarding ERP in patients with congenital LQTS, there are some noteworthy limitations. First, this cohort is mainly white, and it is well established that ethnicity is a significant modulator of ER, with it observed more commonly in black cohorts. Therefore, this may limit the generalizability of our findings. Furthermore, the nature of our institution, our large LQTS practice, and our reputation regarding the evaluation of athletes with genetic heart diseases may contribute to potential referral bias. For example, athletes themselves are more likely to express ERP on the surface ECG. In addition, although beneficial for our patients and their outcomes following diagnosis and implementation of their patient-tailored treatment program, the number of patients who experienced a BCE (n = 39) was small, and, given these therapeutic results, we may not be able to detect a potential prognostic role of ERP in this environment. When examining the dynamics of ERP, we did not take into consideration the number of ECGs in follow-up or have the ability within the current data to examine why these changes may have occurred. Furthermore, we did not manually measure the QTc and relied on automatic QTc calculations. Last, we did not take into account or specifically examine the impact of other modifiable risks after the first ECG, such as left cardiac sympathetic denervation, which could impact the appearance of ERP.
Although early repolarization is common in patients with LQTS, the presence of early repolarization is not associated with either the patient’s symptomatic/asymptomatic classification or the likelihood of a previously symptomatic LQTS patient to experience a BCE following implementation of their treatment program.
COMPETENCY IN MEDICAL KNOWLEDGE: Early repolarization is a common finding in LQTS and, contrary to previous publications, its presence does not seem to be predictive of LQTS symptoms or risk of arrhythmic events.
TRANSLATIONAL OUTLOOK: Although this study reports no association between ERP and LQTS, further studies should be undertaken at a basic science level to examine closely the relationship between Ito and LQTS mutations.
Dr. Ackerman has worked as a consultant and has received equity/royalties from AliveCor, Audentes Therapeutics, Blue Ox Health, Boston Scientific, Gilead Sciences, Invitae, Medtronic, MyoKardia, St. Jude Medicial, and StemoniX. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- breakthrough cardiac events
- early repolarization pattern
- implantable cardioverter defibrillator
- interquartile range
- long QT syndrome
- corrected QT interval
- ventricular fibrillation
- ventricular tachycardia
- Received February 22, 2018.
- Revision received May 30, 2018.
- Accepted June 7, 2018.
- 2018 American College of Cardiology Foundation
- Rohatgi R.K.,
- Sugrue A.,
- Bos J.M.,
- et al.
- Conte G.,
- de Asmundis C.,
- Sieira J.,
- et al.
- Tsuda T.,
- Hayashi K.,
- Konno T.,
- et al.
- Hasegawa K.,
- Watanabe H.,
- Hisamatsu T.,
- et al.
- Macfarlane P.,
- Antzelevitch C.,
- Haissaguerre M.,
- et al.
- Patton K.K.,
- Ellinor P.T.,
- Ezekowitz M.,
- et al.
- Noseworthy P.A.,
- Tikkanen J.T.,
- Porthan K.,
- et al.
- Walsh J.A.,
- Ilkhanoff L.,
- Soliman E.Z.,
- et al.
- Ahmed H.,
- Czosek R.J.,
- Spar D.S.,
- Knilans T.K.,
- Anderson J.B.
- Noseworthy P.A.,
- Weiner R.,
- Kim J.,
- et al.
- Ilkhanoff L.,
- Soliman E.Z.,
- Prineas R.J.,
- et al.