
==== Front
9918367981006676
51512
Int J Cardiol Congenit Heart Dis
Int J Cardiol Congenit Heart Dis
International journal of cardiology. Congenital heart disease
2666-6685

10.1016/j.ijcchd.2022.100408
nihpa2021662
Article
QRS fragmentation versus QRS prolongation in predicting right ventricular enlargement and dysfunction in children and adults with repaired Tetralogy of Fallot
Gaydos Stephanie a*
Hlavacek Anthony a
Evenhouse Susan b
Strelow Jacob c
Chowdhury Shahryar a
Jackson Lanier a
a Division of Pediatric Cardiology, Department of Pediatrics, Shawn Jenkins Children’s Hospital, Medical University of South Carolina, 10 McClennan Banks Drive SJ2190G, Charleston, SC 29425, USA
b Department of Medicine, Medical University of South Carolina, 135 Rutledge Avenue, 12th Floor, MSC Code 591, Charleston, SC 29425, USA
c Texas Center for Pediatric and Congenital Heart Disease, UT Health Austin and Dell Children’s Medical Center, Dell Medical School of the University of Texas at Austin; Dell Pediatric Research Institute 3.812.08, 1400 Barbara Jordan Blvd., Austin, TX 78723, USA
* Corresponding author.: gaydoss@musc.edu (S. Gaydos).
17 9 2024
9 2023
17 6 2022
24 9 2024
13 100408https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Patients with repaired Tetralogy of Fallot (rTOF) have risks of late life-threatening sequelae, including right ventricular (RV) dilation and failure, arrhythmias, and sudden death. QRS prolongation is a well-known ECG predictor of these outcomes but has poor sensitivity for mortality. Growing evidence demonstrates QRS fragmentation (fQRS) as a better prognostic marker for mortality in adults with rTOF, though the two markers have not been directly compared as correlates for CMR abnormalities. Additionally, fQRS has never been studied in pediatric TOF. This single institution retrospectively reviewed 138 CMRs in rTOF patients (median age 21.7 years) who had a corresponding 12-lead ECG within 1 year. fQRS was defined as ≥3 R-waves/notches in the R/S complex (>2 in right bundle branch block) in ≥2 contiguous leads. QRS prolongation was defined as QRS ≥160 ms. Nearly half (46%) the sample had fQRS (42.1% of pediatric subgroup), and 26% had QRS prolongation. Both markers were significantly associated with reduced RV ejection fraction (EF%) (p < 0.01) and larger RV end-diastolic volumes (p < 0.01). QRS prolongation alone predicted lower LV EF% (p = 0.02). Regression analyses showed both QRS prolongation (p < 0.01) and fQRS (p < 0.01) independently associated with reduced RV EF%; QRS prolongation alone predicted RV dilation (p < 0.01). We concluded that both QRS prolongation and fQRS are equivalent as significant markers of RV dysfunction in rTOF patients. QRS prolongation may be a better surrogate for RV dilation specifically. fQRS was frequently seen in children with rTOF and was significantly associated with similar late structural sequelae.

Tetralogy of Fallot
QRS fragmentation
==== Body
pmc1. Introduction

Tetralogy of Fallot (TOF) is the most common cyanotic congenital heart defect and remains one of the most common diagnoses seen in the adult congenital heart disease population [1,2]. Patients born with TOF typically undergo surgical repair early in life and thrive for decades, however, most exhibit residual hemodynamic and electrophysiological sequelae later in life. Risk stratification for these complications, including life-threatening arrhythmia, pulmonic valve dysfunction, right ventricular (RV) dilation, biventricular dysfunction, and death, includes a multitude of patient-specific factors [3–5]. Despite decades of research, patient risk stratification remains subpar in clear prognostication schemes to guide specific medical and procedural interventions. QRS prolongation is a well-known electrocardiogram (ECG) marker associated with some of these clinical outcomes, though it has poor sensitivity for mortality [6,7]. Fragmented QRS (fQRS) is another easily identifiable ECG marker which represents a disruption in conduction due to myocardial fibrosis and scar. It is highly predictive of poor prognosis in patients with coronary artery disease, ischemic and nonischemic forms of cardiomyopathy [8–10]. It has recently been studied in adults with repaired TOF (rTOF) as a marker of right ventricular dilation and dysfunction, fibrosis, right ventricular outflow tract aneurysm, arrhythmia and sudden cardiac death [11–13]. Recent large-scale data demonstrated superiority in fQRS over QRS prolongation in predicting ventricular arrhythmias and mortality in adults with rTOF [6]. These major events remain rare, and outpatient providers rely on periodic imaging studies for surveillance over time. These two ECG markers have not been compared as predictors of late structural abnormalities detected on CMR, such as RV dilation or biventricular dysfunction, which are also strong markers of poor clinical outcomes [7, 14,15]. Additionally, prevalence of fQRS in pediatric TOF patients is unknown. The objective of this study was to compare fQRS with QRS prolongation as markers of CMR abnormalities in adolescents and adults with rTOF.

2. Methods

This was a single institution retrospective chart review of rTOF patients who underwent a CMR at the Medical University of South Carolina between July 15, 2005 and July 15, 2017. Institutional Review Board approval was obtained. Inclusion criteria was patient age ≥10 years at time of CMR and available 12-lead ECG completed within 1 year. This timespan was utilized to capture enough subjects with proximal ECG study to CMR, as ECG is not obtained at every outpatient visit by all providers. A unique study ID was assigned per patient CMR and corresponding ECG, and number of days between CMR and ECG was recorded. The CMR studies were performed using a 1.5 T system (Magnetom Avanto, Siemens Healthineers, Erlangen, Germany), following a standard institutional clinical protocol. The patients were scanned head-first in the supine position. The sessions were initiated with steady state free precession (SSFP) localizing views in the 3 orthogonal planes to determine the position of the ventricles, followed by a 2-chamber localizer, 4-chamber localizer, short-axis localizer, 4-chamber multiphase stack and a short-axis multiphase stack. The multiphase images were retrospectively gated with 25 phases per cardiac cycle and taken during an expiratory breath hold. The field of view was adjusted for body size. The short-axis slices were planned using the 4-chamber multiphase slices with a stack of slices extending from the plane of the atrioventricular valves through the apex (slice thickness 6–8 mm without space between slices). The RV and LV volumes were calculated from the short-axis data set using standard analysis software (Argus, Siemens Healthcare, Erlangen, Germany). The phase of both the end-diastole and end-systole was defined for each LV and RV independently. The phase of the end diastole was defined visually by the observer as the phase with the largest volume. The phase of the end-systole was defined visually by the observer as the phase with the smallest volume. The endocardial contours of the ventricle were traced manually in every slice in which the myocardium of the ventricle was visible. The RV volume was taken as analyzed for the clinical study and was not reanalyzed.

Clinical data such as sex, type of complete TOF surgical repair, year and age at surgical repair, year(s) of pulmonary valve replacement(s), and Holter monitor reports were obtained from electronic medical record. CMR measurements of right and left ventricular end-systolic and end-diastolic volumes indexed to body surface area (RVESVi, RVEDVi, LVESVi, and LVEDVi respectively), right ventricle (RV) ejection fraction (EF%), left ventricle (LV) EF%, and main pulmonary regurgitation fraction were recorded from each CMR report. Other cardiac lesions which can be seen in rTOF patients, including pulmonary valve or branch pulmonic stenosis, residual ventricular septal defects, tricuspid valve regurgitation, were not included in this study. Late gadolinium enhancement protocol was not uniformly performed in all CMR studies within this timeframe, so this data was not analyzed.

All ECGs were analyzed by a single observer, an electrophysiologist, who was blinded to patient characteristics and clinical data. The rhythm, rate, QRS duration, and presence of fQRS in each lead were determined. QRS duration was measured from the earliest to the latest signal crossing the isoelectric line of the broadest QRS in any lead. QRS duration ≥160 ms was defined as prolonged. In patients with wide complex QRS (≥120 ms), right bundle branch block, or paced QRS complexes, fQRS was defined as ≥3 R-waves/notches in the R/S complex (more than typical 2 in right bundle branch block) in ≥2 contiguous leads (Fig. 1). In patients with QRS duration <120 ms, fragmentation was defined as an additional R wave (R’) or notch in the nadir of the R wave or the S wave. Severity was qualified as mild if there were <4 contiguous leads with fQRS and moderate-severe if present in ≥4 leads. These definitions were consistent with previously reported protocols [6, 8,9,11]. There was 99.3% intra-observer agreement for the presence of fQRS. The most proximal Holter monitor to time of CMR was reviewed with documentation of major findings for each subject ID.

The primary outcome variable was right ventricular dilation by CMR volume quantification (indexed to body surface area). Secondary outcomes included RV function (CMR quantified RV EF%), LV function (LV EF%), pulmonary regurgitation fraction, and non-sustained or sustained ventricular tachycardia on available Holter reports (non-sustained defined as 3 or more consecutive ventricular beats at a rate exceeding 100 beats per minute and lasting <30 s). fQRS and QRS prolongation were separately analyzed for associations with these outcomes in both the entire cohort and separately in a pediatric subgroup (subjects age ≤18 years). Associations between fQRS and years since initial surgical repair as well as TOF surgical type were evaluated. Subject characteristics were described using medians and interquartile ranges (IQR) when appropriate. Continuous variables were presented as means and analyzed with ANOVA, Fisher’s exact test, and linear regression models. Analyses were performed with IBM SPSS Statistics software v. 24 (manufactured in Armonk, NY). A p-value <0.05 was considered statistically significant.

3. Results

A total of 162 CMR reports of patients age ≥10 years with rTOF were identified. Of these, there were 138 corresponding 12-lead ECGs performed within 1 year of the CMR available to review. Median age at CMR was 21.7 years (IQR 15.4, 30.7) with 41% of subjects age ≤18 years. Table 1 summarizes the baseline clinical characteristics of these 138 subjects, as well as CMR and 12-lead ECG data. The majority of the cohort underwent a transannular patch (n = 102, 76%), and the median time from complete TOF repair and CMR was 19.0 years (IQR 15.0, 28.8). There was documented history of pulmonary valve replacement in 79 subjects (57%), similar to prevalence in half of all patients by age 30 years in other series; however, this is likely an underestimate given incomplete history to review in all subjects.

Nearly half of the cohort (n = 64, 46%) had fQRS on 12-lead ECG near time of CMR. Of these, 78% exhibited mild fragmentation in <4 leads (n = 50) while 22% (n = 14) had moderate-to-severe fragmentation in ≥4 leads. Fragmented QRS was predominantly noted in the anteroseptal segments (leads V1–V5), followed in by inferior segments (leads II, III, aVF) and infrequently seen in lateral segment leads (V6, I, aVL) (Fig. 2). Presence of fQRS was not significantly associated with complete surgical repair type (p = 0.98) nor duration since repair (p = 0.10). Only 26% of the cohort (n = 36) met the definition of prolonged QRS duration, with 6 subjects exhibiting severe QRS prolongation (QRS ≥180 ms). Both ECG markers were simultaneously found in 19% of the sample (n = 26). There was a corresponding Holter monitor report obtained within approximately 1 year of CMR in 73 subjects (53%). Non-sustained ventricular tachycardia was documented on 2 subjects’ reports; one of these subjects had moderate-severity fQRS, and the other had neither fQRS nor prolonged QRS on corresponding ECG. There was no mortality in the cohort during the time period of review.

Univariable linear regression was used to test if each ECG marker significantly predicted biventricular function and RVEDVi. Both fQRS and QRS prolongation significantly predicted reduced RV EF% with similar strength (R2 = 0.175 and p < 0.01for each). Only QRS prolongation significantly predicted greater RVEDVi (p < 0.001) and reduced LV EF% (p = 0.02). Multivariable analyses showed the presence of fQRS to be independently associated with increased RVEDVi (p = 0.01) and RVESVi (p < 0.01), as well as reduced RV systolic function (p < 0.01) compared to those without this ECG marker (Table 2, Fig. 3). Severity of fQRS was not correlated with meaningful differences. Similarly, QRS prolongation was independently associated with increased RV systolic and diastolic volumes and reduced RV EF% (p < 0.01) compared to those normal QRS duration, as well as an independent association with reduced LV EF% (p = 0.02). Neither the presence of fQRS or QRS prolongation was correlated with significant differences in pulmonary regurgitation fraction. Subjects with prolonged QRS were older at time of CMR with greater duration since their surgical repair compared to those with normal QRS duration; this trend was not seen with respect to fQRS presence or absence.

A subgroup of 57 pediatric subjects was identified by age ≤18 years at time of CMR. The median age of complete TOF surgical repair in this group was 0.7 years (IQR 0.25, 0.9) with the majority undergoing transannular patch (n = 40, 72.7%) (Table 3). fQRS was present in 24 subjects’ ECGs (42%), and 6 of these had moderate-severe fractionation. Only 8 subjects (14%) had QRS prolongation, with a shorter duration from complete repair (median 13.0 years) compared to those >18 years of age (median 20.0 years from complete repair, 28 subjects (35%) with QRS prolongation). Univariable linear regression tests demonstrated that fQRS was significantly associated with RV dilation by larger RVEDVi (R2 = 0.10, p = 0.02) and RVESVi (R2 = 0.18, p < 0.01), reduced RV EF% (R2 = 0.17, p < 0.01), and reduced LV EF% (R2 = 0.13, p < 0.01). Multivariable analyses also showed fQRS as independently associated with these outcomes: larger RVEDVi (p = 0.02), larger RVESVi (p < 0.01), reduced RV EF% (p < 0.01), and reduced LV EF% (p < 0.01).

4. Discussion

Surveillance of long-term clinical sequelae in TOF patients following repair is an evolving challenge in pediatric and adult congenital cardiology, with significant risks of residual pulmonary valve disease, heart failure, arrhythmia, and sudden cardiac death. This is an important task identified as a high-priority research topic by the National Heart, Lung, and Blood Institute/Adult Congenital Heart Association Working group in 2016 [16]. In the search for best markers and algorithms for patient risk prediction, 12-lead ECG and CMR are two highly utilized diagnostic tools in clinical decision pathways. While there is ongoing investigation to understand how these various electrical and structural findings relate to one another, there has not been direct comparison of two specific ECG markers of adverse outcomes, QRS fractionation and QRS prolongation, to CMR abnormalities in rTOF patients. In this study of 138 CMR’s of adolescents and adults with rTOF, we found that both ECG markers were independently predictive of reduced RV systolic function. Only QRS prolongation was significantly associated with other CMR abnormalities including RV enlargement and LV systolic dysfunction. These findings also offer the first description of QRS fragmentation prevalence in children after TOF repair with similarly significant associations with CMR abnormalities.

Currently, CMR has become the gold standard for accurate structural assessments of biventricular size, function, fibrosis, and valvular regurgitation in rTOF patients. CMR measurements of RV systolic and diastolic size and function have subsequently factored into criteria for pulmonary valve replacement guidelines and are also found to be associated with risks of sudden cardiac death and other adverse outcomes [4,14,17,18]. Serial CMR’s to trend these quantifiable data are common practice in caring for patients with rTOF today to help guide clinical decision-making in conjunction with symptoms and other multimodality assessments. However, there are frequently practical challenges in serially obtaining appropriate CMRs in this patient group. Depending on geography, it may be challenging to schedule a CMR at a congenital cardiac center equipped with necessary components for the image acquisition protocol and for subsequent interpretation in rTOF patients. Issues obtaining insurance approval are common, as is image interference from metallic cardiac devices such as pacemaker leads or pulmonary artery stents. Lastly, providers frequently encounter patient hesitance related to anxiety or claustrophobia with a lengthy scan.

The 12-lead ECG is a valuable and clinically feasible tool in routine assessments of TOF patients. ECGs initially offered foundational data with QRS duration ≥180 ms showing association with adverse outcomes, though this marker has since found to be more limited than initially suspected in sensitivity to predict mortality [7,14,19,20]. Attention has shifted to a new electrocardiographic marker: QRS fragmentation, a surrogate for myocardial fibrosis. This abnormal ECG marker is most widely studied in non-congenital cardiac pathologies such as coronary artery disease and cardiomyopathies, been found to be an important marker the adult congenital heart disease population. There has been growing interest among adults with rTOF given the increasing population of these surviving patients, propensity for arrhythmia, and the known myocardial fibrosis following various surgical repairs. Bokma et al. recently highlighted fQRS as a superior predictor of mortality compared to QRS prolongation in adults with rTOF, which was validated by Egbe et al. [6,21].

Unlike CMR, an ECG can be easily performed in any cardiologist’s office and is relatively easily interpretable. While ECG markers such as fQRS and QRS prolongation cannot be used in isolation to estimate broad prognosis and clinical decision-making for complex patients such as those after repair of TOF, they can certainly contribute to the broad picture during annual evaluation. Our findings support that identification of either ECG marker can potentially serve as a trigger to prompt further investigation and treatment of cardiac structural sequelae such as right ventricular dilation or dysfunction, with new attention to significance of fQRS even in children with rTOF. Specifically, these markers could be considered in mapping out a patient’s indication or timing for CMR to optimize resource utilization in long-term monitoring. Many clinicians have used a combination of non-invasively derived ECG markers such as these plus induction of sustained ventricular tachyarrhythmia on programmed ventricular stimulation as part of patient risk-stratification of whether an implantable defibrillator is indicated, and this could also be considered with these markers’ significance [22].

While this study adds to the knowledge about the relationship between these electrical markers and structural changes in rTOF patients, it also highlights areas needing further study. Residual lesions commonly seen in rTOF patients, including pulmonary valve regurgitation or stenosis, tricuspid regurgitation, residual ventricular septal defect, and/or right ventricular outflow tract aneurysm, can impact cardiac structure and hemodynamics, and hence some of these lesions are associated with QRS abnormalities. While there was no relationship between pulmonary regurgitation and either ECG marker in these findings, it may remain a question given missing CMR data; other studies have shown both presence and absence of an association with fQRS [11, 12]. Additional work is needed to better understand the potential impact of these lesions. The significance of fQRS presence in pediatric TOF patients remains unknown, as prior studies investigating adverse cardiovascular events excluded those under 18 years of age. We now see that fQRS is prevalent even in young patients, and additional research is needed to answer whether there is similar association with poor clinical outcomes. The association between fQRS with LV dysfunction seen only in this younger subgroup should specifically be further studied. If reproducible, this finding could be explained by greater ventricular fibrosis related to surgical repair at younger ages, including ventricular septal defect patch size which may impact left ventricular function if resulting in a large area of dyskinesis. Additionally, the time-related changes in fQRS and QRS prolongation with interval CMR changes needs to be elucidated. There remains a question of whether interventions such as pulmonary valve replacement result in similar electrocardiographic changes as structural ones from cardiac remodeling, and whether clinical outcomes change with them. Understanding these chronological and interventional influences will facilitate better utilization of each ECG marker when considering workup for an intervention or in discussions about future prognosis.

Interestingly, there was not a trend of non-sustained ventricular tachycardia or other ectopy burden on available Holter monitors with QRS fragmentation or QRS prolongation, as one may have expected given the associations with these arrhythmogenic outcomes in existing studies. This may be explained by the overall low prevalence of malignant arrhythmias in TOF survivors and younger age of this cohort. However, in a few similar study populations of rTOF patients, these associations were seen specifically with abnormal signal-averaged ECG markers, whereas QRS duration was not [23,24]. There may be interest in a direct comparison between signal-averaged ECG indices with QRS fragmentation in the future regarding some of these clinical outcomes, including CMR findings as was done in this study.

This study was limited to single-center retrospective review of existing CMRs, with small sample for robust subgroup analysis. A larger sample may have enough power to analyze whether more severe QRS prolongation (≥180 ms, a cutoff with well-known associated clinical outcomes) has the same associations with structural abnormalities found in this study using a less severe QRS duration cutoff. There was variability in CMR imaging protocol over the study timespan (ie, contrast protocol or consistent documentation of pulmonary regurgitation fraction). Quantifications were taken from documented CMR reports rather than repeat interpretation by a single independent observer for the purposes of this study. It would be valuable to compare these ECG markers with respect to late gadolinium enhancement, extracellular volume and T1 mapping, given the known association of QRS fragmentation in particular with fibrosis, and this could be incorporated into a study protocol for future work. Clinical outcomes were examined as secondary endpoints and limited to documented death or arrhythmia on event monitor, which does not capture all relevant cardiovascular adverse effects of interest–particularly in the pediatric subgroup whom represent an unstudied group within this area. Lastly, the impact of pulmonary valve replacement with regards to timing of ECG and CMR was not able to be unanimously accounted for in chart review, as subjects’ complete clinical and interventional histories were not uniformly available.

5. Conclusions

These study results demonstrate that a standard 12-lead ECG can provide readily accessible risk stratification in routine follow-up of children and adults with rTOF, which can complement other clinical data to trigger timing of additional investigations. Both QRS prolongation and fQRS are significant markers of RV dysfunction in adolescents and adults with rTOF; they seem equivalent by this comparison. QRS prolongation alone was found to be a predictor of RV dilation and LV dysfunction, whereasfQRS was frequently seen in children with rTOF with the the potential to predict the same CMR trends. Our Our findings should be examined for relevance to clinicalrelevance outcomes in future, prospective and larger studies in this patient group.future future

Abbreviations

TOF Tetralogy of Fallot

rTOF repaired Tetralogy of Fallot

ECG electrocardiogram

fQRS QRS fragmentation

CMR cardiac magnetic resonance imaging

RV right ventricle

LV left ventricle

EF ejection fraction

EDVi end-diastolic volume indexed to body surface area

ESVi end-systolic volume indexed to body surface area

Fig. 1. Example of QRS fragmentation.

Patient with repaired Tetralogy of Fallot and right bundle branch block with fragmented QRS complexes in leads V2, V3, V4, and V5. Fragmentation is defined as characteristic 3 notches in the R/S complex in the presence of a right bundle branch block, a common finding in rTOF patients.

Fig. 2. Distribution of fQRS complexes in entire cohort by ECG lead segments.

Distribution of fQRS complexes by individual ECG lead segments. Leads V1–V5 represent anteroseptal region; leads V6, I, aVL represent lateral segments; and leads II, III, aVF represent inferior region.

Fig. 3. Cardiac MRI trends with fQRS and prolonged QRS duration in patients with repaired Tetralogy of Fallot.

Multivariable analyses of ECG markers with ventricular abnormalities on CMR. Left graph shows fQRS and QRS prolongation, “QRSp,” to be independently associated with increased RVEDVi. (p = 0.01 and p < 0.01, respectively). Middle graph shows fQRS and QRSp as each independently associated with reduced RV EF% (p < 0.01 for each marker). Right graph shows that fQRS was not significantly associated with differences in LV EF%. The presence of QRSp was independently associated with reduced LV EF% (p = 0.02).

Table 1 Clinical characteristics of the study population.

	Entire cohort (n = 138)	
	
Males	57 (41.3%)	
Age at complete TOF surgical repair	1.3 (0.7, 3.0)	
Type of surgical repair		
 Transannular patch	102 (75.6%)	
 Conduit	19 (14.1%)	
 Valve-sparing	14 (10.4%)	
Age at CMR	21.6 (15.4, 30.7)	
Years between CMR and surgical repair	19.0 (15.0, 29.0)	
Days between CMR and ECG	52.5 (18.3, 114.5)	
History of pulmonary valve replacement	79 (57%)	
CMR data		
 RVEDVi (ml/m2)	122.0 (102.4, 152.1)	
 RVESVi (ml/m2)	62.5 (43.9, 82.1)	
 RVEF (%)	51.0% (±10.4)	
 LVEDVi (ml/m2)	69.5 (62.2, 79.0)	
 LVESVi (ml/m2)	28.4 (22.5, 34,4)	
 LVEF (%)	59.3% (±8.0)	
 Pulmonary regurgitation fraction (%)	28.4% (±16.2)	
ECG data		
 QRS duration (ms)	144 (124, 160)	
Time between ECG and CMR (days)	52.5 (18.3, 111.5)	
Data are expressed as median (interquartile range), means (± standard deviation), or as a number (percentage).

Note: Missing data for surgical repair type in 3 subjects, and missing pulmonic regurgitation fraction in 32 CMR reports.

Table 2 Multivariate analyses of fQRS presence and prolonged QRSd with structural findings on CMR in patients with repaired Tetralogy of Fallot. CMR measurements reported as mean ± standard deviation.

	fQRS absent (n = 74)	fQRS present (n = 64)	p-value	Normal QRSd (n = 102)	Prolonged QRSd (n = 36)	p-value	
	
Males	27 (36%)	30 (47%)	0.21	35 (34%)	22 (61%)	<0.01	
Age at repair (years)	1.8 ± 2.3	3.9 ± 8.2	0.05	1.8 ± 2.6	5.4 ± 10.4	0.05	
Age at MRI (years)	22.9 ± 9.7	26 ± 13.1	0.07	22.3 ± 9.8	31.0 ± 13.5	<0.01	
Years between MRI and surgical repair	21 ± 8.7	23.8 ± 10.6	0.10	20.6 ± 8.5	27.1 ± 11.3	<0.01	
Days between MRI and ECG	88.0 ± 96.5	84.9 ± 97.9	0.86	95.8 ± 99.2	60.3 ± 85.7	0.04	
RVEDVi (ml/m2)	121.0 ± 35.7	138.2 ± 39.7	0.01	121.5 ± 31.4	148.9 ± 48.1	<0.01	
RVESVi (ml/m2)	57.3 ± 26.7	75.5 ± 31.4	<0.01	58.5 ± 22.6	84.8 ± 39.2	<0.01	
RVEF(%)	54.1% ± 9.5%	47.1% ± 10.3%	<0.01	53.2% ± 9.0%	44.9% ± 11.9%	<0.01	
LVEDVi (ml/m2)	71.6 ± 16.6	73.8 ± 9.0	0.51	72.3 ± 17.1	73.5 ± 21.8	0.74	
LVESVi (ml/m2)	28.8 ± 10.3	30.7 ± 9.1	0.28	28.7 ± 8.3	32.5 ± 12.9	0.06	
LVEF (%)	60.3% ± 7.4%	58.0% ± 8.5%	0.11	60.2% ± 7.4%	56.6% ± 8.9%	0.02	
Pulmonary regurgitation fraction	29.2% ± 15.5%	27.5% ± 17.4%	0.61	27.4 ± 15.7%	31.5% ± 18.5%	0.28	

Table 3 Pediatric subgroup describes characteristics of subjects ages ≤18 years at time of MRI. Data are expressed as median (interquartile range), means (± standard deviation), or as a number (percentage).

Characteristics of pediatric subgroup	Total n = 57	
	
Males	28 (49.1%)	
Age at complete TOF surgical repair	0.7 (0.25, 0.9)	
Type of surgical repair		
 Transannular patch	40 (72.7%)	
 Conduit	10 (18.2%)	
 Valve-sparing	5 (9.1%)	
History of pulmonary valve replacement	30 (52.6%)	
Age at CMR	14.2 (12.5, 16.6)	
Years between CMR and surgical repair	13.0 (12.0, 16.0)	
Days between CMR and ECG	58.0 (21.0, 157.0)	
CMR data		
 RVEDVi (ml/m2)	119.0 (98.8, 146)	
 RVESVi (ml/m2)	62.1 (37.9, 81.0)	
 RVEF (%)	52.7% ± 0.1	
 LVEF (%)	60.7% ± 0.1	
Pulmonary regurgitant fraction (%)	28.5% ± 0.2	
ECG data		
 QRS duration (ms)	134.6 ± 26.8	
 fQRS (n, %)	24 (42%)	
 QRS prolongation (n, %)	8 (14%)	
Note: Missing data for surgical repair type in 2 subjects in pediatric cohort, and missing pulmonic regurgitation fraction in 15 CMR reports.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Disclosures

Authors have no disclosures.
==== Refs
References

[1] Wu MH , Lu CW , Chen HC , Kao FY , Huang SK . Adult congenital heart disease in a nationwide population 2000–2014: epidemiological trends, arrhythmia, and standardized mortality ratio. J Am Heart Assoc 2018;7 (4 ).
[2] Brickner ME , Hillis LD , Lange RA . Congenital heart disease in adults. Second of two parts. N Engl J Med 2000;342 (5 ):334–42.10655533
[3] Khairy P , Aboulhosn J , Gurvitz MZ , Opotowsky AR , Mongeon FP , Kay J , Arrhythmia burden in adults with surgically repaired tetralogy of Fallot: a multi-institutional study. Circulation 2010;122 (9 ):868–75.20713900
[4] Geva T , Sandweiss BM , Gauvreau K , Lock JE , Powell AJ . Factors associated with impaired clinical status in long-term survivors of tetralogy of Fallot repair evaluated by magnetic resonance imaging. J Am Coll Cardiol 2004;43 (6 ):1068–74.15028368
[5] Murphy JG , Gersh BJ , Mair DD , Fuster V , McGoon MD , Ilstrup DM , Long-term outcome in patients undergoing surgical repair of tetralogy of Fallot. N Engl J Med 1993;329 (9 ):593–9.7688102
[6] Bokma Jouke PWMM , Vehmeijer Jim T , Vliegen HW , van Dijk AP , van Melle JP , Meijboom FJ , Post MC , Zwinderman AH , Mulder BJM . QRS fragmentation is superior to QRS duration in predicting mortality in adults with tetralogy of Fallot. Heart 2017;(103 ):666–71.27803032
[7] Gatzoulis MA , Till JA , Somerville J , Redington AN . Mechanoelectrical interaction in tetralogy of Fallot. QRS prolongation relates to right ventricular size and predicts malignant ventricular arrhythmias and sudden death. Circulation 1995;92 (2 ):231–7.7600655
[8] Jain R , Singh R , Yamini S , Das MK . Fragmented ECG as a risk marker in cardiovascular diseases. Curr Cardiol Rev 2014;10 (3 ):277–86.24827794
[9] Das MK , Zipes DP . Fragmented QRS: a predictor of mortality and sudden cardiac death. Heart Rhythm 2009;6 (3 Suppl ):S8–14.
[10] Cho MJ , Lee JW , Lee J , Shin YB , Lee HD . Relationship between fragmented QRS complexes and cardiac status in duchenne muscular dystrophy: multimodal validation using echocardiography, magnetic resonance imaging, and holter monitoring. Pediatr Cardiol 2017;38 (5 ):1042–8.28456833
[11] Nesan Shanmugan JY , Tan Ru San , Le Thu Thao , Gao Fei , Chan Jia Xin , Chong Daniel , Leng Ho Kah , Yew Tan Boon , Keong Ching Chi , Teo Wee Siong , Tan Ju Le , Liew Reginald . Fragmented QRS complexes predict right ventricular dysfunction and outflow tract aneurysms in patients with repaired tetralog of Fallot. Int J Cardiol 2012;167 (4 ):1366–72.22521381
[12] Seung-Jung Park YKO , Kim June Soo , Park Seung Woo , Yang Ji-Hyuk , Jun Tae-Gook , Kang I-Seok , Lee Heung Jae , Choe Yeon Hyeon , Huh June . Relation of fragmented QRS complex to right ventricular fibrosis detected by late gadolinum enhancement cardiac magnetic resonance in adults with repaired tetralogy of fallot. Am J Cardiol 2012;109 (1 ):110–5.21962997
[13] Vehmeijer JT , Koyak Z , Bokma JP , Budts W , Harris L , Mulder BJM , Sudden cardiac death in adults with congenital heart disease: does QRS-complex fragmentation discriminate in structurally abnormal hearts? Europace 2018;20 (Fi1 ):f122–8.28402450
[14] Valente AM , Gauvreau K , Assenza GE , Babu-Narayan SV , Schreier J , Gatzoulis MA , Contemporary predictors of death and sustained ventricular tachycardia in patients with repaired tetralogy of Fallot enrolled in the INDICATOR cohort. Heart 2014;100 (3 ):247–53.24179163
[15] Possner M , Tseng SY , Alahdab F , Bokma JP , Lubert AM , Khairy P , Risk factors for mortality and ventricular tachycardia in patients with repaired tetralogy of fallot: a systematic review and meta-analysis. Can J Cardiol 2020;36 (11 ):1815–25.32416063
[16] Gurvitz M , Burns KM , Brindis R , Broberg CS , Daniels CJ , Fuller SM , Emerging research directions in adult congenital heart disease: a report from an NHLBI/ACHA working group. J Am Coll Cardiol 2016;67 (16 ):1956–64.27102511
[17] Stout KK , Daniels CJ , Aboulhosn JA , Bozkurt B , Broberg CS , Colman JM , AHA/ACC guideline for the management of adults with congenital heart disease: a report of the American college of cardiology/American heart association task force on clinical practice guidelines. J Am Coll Cardiol 2018;73 (12 ):e81–192. 2019. 30121239
[18] Geva T Repaired tetralogy of Fallot: the roles of cardiovascular magnetic resonance in evaluating pathophysiology and for pulmonary valve replacement decision support. J Cardiovasc Magn Reson 2011;13 (1 ):9.21251297
[19] Müller J , Hager A , Diller GP , Derrick G , Buys R , Dubowy KO , Peak oxygen uptake, ventilatory efficiency and QRS-duration predict event free survival in patients late after surgical repair of tetralogy of Fallot. Int J Cardiol 2015;196 :158–64.26114442
[20] Gatzoulis MA , Balaji S , Webber SA , Siu SC , Hokanson JS , Poile C , Risk factors for arrhythmia and sudden cardiac death late after repair of tetralogy of Fallot: a multicentre study. Lancet 2000;356 (9234 ):975–81.11041398
[21] Egbe AC , Miranda WR , Mehra N , Ammash NM , Missula VR , Madhavan M , Role of QRS fragmentation for risk stratification in adults with tetralogy of fallot. J Am Heart Assoc 2018;7 (24 ):e010274.30561260
[22] Gatzoulis K , Frogoudaki SB , Stefanadis C . Implantable defibrillators: from the adult cardiac to the grown up congenital heart disease patient. Int J Cardiol 2004;97 (1 ):117–22.15590088
[23] Brili S , Aggeli C , Gatzoulis K , Tzonou A , Hatzos C , Pitsavos C , Stefanadis C , Toutouzas P . Echocardiographic and signal averaged ECG indices associated with non-sustained ventricular tachycardia after repair of tetralogy of fallot. Heart 2001 Jan;85 (1 ):57–60.11119464
[24] Russo G , Folino AF , Mazzotti E , Rebellato L , Daliento L . Comparison between QRS duration at standard ECG and signal-averaging ECG for arrhythmic risk stratification after surgical repair of tetralogy of fallot. J Cardiovasc Electrophysiol 2005 Mar;16 (3 ):288–92.15817088
