Author + information
- Received May 18, 2017
- Revision received June 13, 2017
- Accepted June 15, 2017
- Published online August 7, 2017.
- Fabien Labombarda, MDa,
- Robert Hamilton, MDb,
- Azadeh Shohoudi, PhDc,
- Jamil Aboulhosn, MDd,
- Craig S. Broberg, MDe,
- Marie A. Chaix, MDa,
- Scott Cohen, MDf,
- Stephen Cook, MDg,
- Annie Dore, MDa,
- Susan M. Fernandes, LPD, PA-Ch,
- Anne Fournier, MDi,
- Joseph Kay, MDj,
- Laurent Macle, MDa,
- Blandine Mondésert, MDa,
- François-Pierre Mongeon, MD, SMa,
- Alexander R. Opotowsky, MD, MMSck,
- Anna Proietti, RNa,
- Lena Rivard, MDa,
- Jennifer Ting, MDl,
- Bernard Thibault, MDa,
- Ali Zaidi, MDm,
- Paul Khairy, MD, PhDa,c,∗ (, )
- on behalf of the AARCC
- aMontreal Heart Institute, Université de Montréal, Montreal, Quebec, Canada
- bThe Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada
- cMontreal Health Innovations Coordinating Center (MHICC), Montreal, Quebec, Canada
- dAhmanson/UCLA Adult Congenital Heart Disease Center, University of California, Los Angeles, Los Angeles, California
- eOregon Health and Science University, Portland, Oregon
- fThe Wisconsin Adult Congenital Heart (WAtCH) Program, Medical College of Wisconsin, Milwaukee, Wisconsin
- gChildren's Hospital of Pittsburgh, Pittsburgh, Pennsylvania
- hDepartments of Pediatrics and Medicine, Stanford University, Palo Alto, California
- iCHU mère-enfant Sainte-Justine, Université de Montréal, Montreal, Quebec, Canada
- jUniversity of Colorado Denver, Aurora, Colorado
- kBoston Adult Congenital Heart Service, Boston Children's Hospital and Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts
- lMilton S. Hershey Medical Center, Pennsylvania State University, Hershey, Pennsylvania
- mNationwide Children's Hospital, Ohio State University, Columbus, Ohio
- ↵∗Address for correspondence:
Dr. Paul Khairy, Montreal Heart Institute Adult Congenital Center, Montreal Heart Institute, 5000 Belanger Street East, Montreal, Quebec H1T 1C8, Canada.
Background Atrial arrhythmias are the most common complication encountered in the growing and aging population with congenital heart disease.
Objectives This study sought to assess the types and patterns of atrial arrhythmias, associated factors, and age-related trends.
Methods A multicenter cohort study enrolled 482 patients with congenital heart disease and atrial arrhythmias, age 32.0 ± 18.0 years, 45.2% female, from 12 North American centers. Qualifying arrhythmias were classified by a blinded adjudicating committee.
Results The most common presenting arrhythmia was intra-atrial re-entrant tachycardia (IART) (61.6%), followed by atrial fibrillation (28.8%), and focal atrial tachycardia (9.5%). The proportion of arrhythmias due to IART increased with congenital heart disease complexity from 47.2% to 62.1% to 67.0% in patients with simple, moderate, and complex defects, respectively (p = 0.0013). Atrial fibrillation increased with age to surpass IART as the most common arrhythmia in those ≥50 years of age (51.2% vs. 44.2%; p < 0.0001). Older age (odds ratio [OR]: 1.024 per year; 95% confidence interval [CI]: 1.010 to 1.039; p = 0.001) and hypertension (OR: 2.00; 95% CI: 1.08 to 3.71; p = 0.029) were independently associated with atrial fibrillation. During a mean follow-up of 11.3 ± 9.4 years, the predominant arrhythmia pattern was paroxysmal in 62.3%, persistent in 28.2%, and permanent in 9.5%. Permanent atrial arrhythmias increased with age from 3.1% to 22.6% in patients <20 years to ≥50 years, respectively (p < 0.0001).
Conclusions IART is the most common presenting atrial arrhythmia in patients with congenital heart disease, with a predominantly paroxysmal pattern. However, atrial fibrillation increases in prevalence and atrial arrhythmias progressively become permanent as the population ages.
- cohort studies
- congenital heart defects
- intra-atrial re-entrant tachycardia
Improved survival of children born with heart defects has resulted in a rapidly expanding population with congenital heart disease (1,2). As these patients age, they face numerous potential long-term complications, of which arrhythmias figure prominently (3). Indeed, atrial arrhythmias are the leading cause of morbidity and hospital admissions (4,5), and have been implicated as a cause of sudden death (6). Although the high burden of arrhythmias is well established in the population with congenital heart disease, little is known about changing types and patterns of arrhythmias as patients age. Detailed prevalence studies on arrhythmias have primarily focused on populations with specific congenital heart defects (7–9), whereas larger studies based on administrative databases have lacked the granularity to distinguish between various types of arrhythmias and their clinical patterns (5). We therefore undertook a multicenter study with detailed characterization of atrial arrhythmias in order to shed light on the types and patterns encountered and age-related trends.
The study protocol was previously described (10). In brief, the study population consisted of patients with heterogeneous forms of congenital heart disease born before 2003, with documented sustained (≥30 s) atrial tachyarrhythmia or atrial fibrillation. Time of entry was defined as the time of the first electrocardiographically documented episode of sustained atrial tachycardia or atrial fibrillation (i.e., by 12-lead electrocardiogram, rhythm strip, ambulatory electrocardiography recording, or cardiovascular implantable electronic device). All eligible patients were systematically recruited from 12 centers across North America (3 in Canada, 9 in the United States) via the Alliance for Adult Research in Congenital Cardiology (AARCC). Data collection was performed between January 1, 2013, and March 1, 2015. During this time frame, case report forms were completed, copies of all required original documents were collected and submitted for independent adjudication, and responses were provided to data quality control measures.
Atrial tachycardia was defined as a regular atrial rhythm of abrupt onset, with a constant rate ≥100 beats/min, originating outside the region of the sinus node (11). As the study focused on long-term ramifications of structural congenital heart disease and surgical interventions, arrhythmias such as atrioventricular nodal re-entrant tachycardia, accessory pathway-mediated tachycardia, automatic ectopic atrial tachycardia, and junctional tachycardia did not qualify for entry. Arrhythmias were classified as focal atrial tachycardia (FAT) if well-formed P waves were distinguishable electrocardiographically and separated by an isoelectric interval, or if an electrophysiology study demonstrated a circumscribed origin of activation with centrifugal spread. Otherwise, the atrial tachycardia was classified as intra-atrial re-entrant tachycardia (IART), a term that encompasses all forms of typical and atypical atrial flutter (11). Atrial fibrillation was defined as the absence of consistent P waves, with rapid oscillations or fibrillatory waves that varied in size, shape, and timing, and were associated with an irregular ventricular response when atrioventricular conduction was intact (12). Data were systemically collected on these 3 types of arrhythmias, including date of first diagnosis and denominalized copies of documented arrhythmias and electrophysiological studies, if applicable, for adjudication.
The predominant clinical pattern of atrial arrhythmias was classified according to standard criteria as paroxysmal (i.e., spontaneous termination <7 days), persistent (i.e., lasting >7 days with spontaneous termination or terminated by cardioversion), or permanent (i.e., failed cardioversion or no attempt to terminate, or with termination, but with subsequent relapse and decision to abandon attempts to terminate) (13).
A multicenter retrospective cohort study was coordinated by the Montreal Health Innovations Coordinating Center (MHICC), which oversaw data collection, data integration, data entry, quality edit checks, management and resolution of data discrepancies, and analyses. Data quality control consisted of a double data-entry process; procedures for flagging illegible data, invalid formats, and invalid codes; systematic range and consistency checks; and inspection by an independent, internal quality-control group, with full review of a randomly selected sample of √n + 1 subjects (where n = total study group), comparing electronic data to case report forms and to corrections provided by data clarification forms. The acceptable error rate was predefined as <0.5%. The observed error rate before database lock was 0.02%.
Data were extracted from existing electronic databases, medical and surgical records, and chart review. Baseline characteristics at the time of the qualifying arrhythmia included age; sex; body mass index; type and complexity of congenital heart disease; history of coronary artery disease or stroke; systemic ventricular ejection fraction; New York Heart Association (NYHA) functional class; resting oxygen saturation; presence of a pacemaker or implantable cardioverter-defibrillator; and hemoglobin, hematocrit, serum sodium, and serum creatinine levels. Congenital heart disease complexity was categorized according to a previously proposed classification scheme as simple, moderate, or complex (14).
A blinded adjudicating committee consisting of 4 of the authors (L.M., F.P.M., L.R., and B.T.) reviewed all qualifying atrial arrhythmias and deaths. Discrepancies were subject to committee discussions for final adjudication. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki, as reflected in a priori approval by each institution's human research committee, with waiver of informed consent. The study adhered to the principles outlined by the International Council of Harmonization Tripartite Guidelines for Good Clinical Practice.
Continuous variables are presented as mean ± SD or median and interquartile range (25th to 75th percentile), depending on normality of distribution. Categorical variables are summarized by frequencies and percentages. Comparisons of categorical baseline characteristics across 3 different types of arrhythmias were performed by chi-square or Fisher exact tests, as appropriate. Continuous characteristics were compared using univariable nominal logistic regression analyses by modeling arrhythmia type as the outcome. A stepwise variable selection approach was conducted to select the final multivariable nominal logistic regression model for types of arrhythmias. Following verification of proportionality assumptions, factors associated with new-onset atrial fibrillation (i.e., in the subgroup of patients without atrial fibrillation at study entry) were assessed in univariable and stepwise multivariable Cox regression analyses, with censoring at the time of last follow-up. Baseline characteristics listed in Table 1 that were associated with p values <0.25 in univariable analyses were considered in multivariable models. Two Cox regression approaches were considered. The first consisted of censoring at time of death. The second considered death a competing risk (i.e., cause-specific hazard model) for new-onset atrial fibrillation. In keeping with the principle of parsimony, the first model was retained, as the 2 approaches yielded similar results. Two-tailed p values <0.05 were considered statistically significant. All analyses were performed using SAS software version 9.4 (SAS Institute, Cary, North Carolina).
A total of 482 patients, 54.8% female, with congenital heart disease and sustained atrial tachyarrhythmias were enrolled. Baseline characteristics overall and according to type of atrial arrhythmia are summarized in Table 1. Mean age at the time of the presenting arrhythmia was 32.0 ± 18.0 years. Type of congenital heart disease was classified as simple in 18.5%, moderate in 34.4%, and complex in 47.1%. Most (91.1%) patients had NYHA functional class I or II symptoms, with a mean resting oxygen saturation of 94.3 ± 7.8%. Pacemakers were implanted in 37.3% and implantable cardioverter-defibrillators in 7.3%. The most prevalent comorbidities were hypertension (14.3%), congestive heart failure (7.5%), dyslipidemia (6.4%), and diabetes (4.8%).
Characteristics according to type of atrial arrhythmia
The most common presenting arrhythmia was IART (61.6%), followed by atrial fibrillation (28.8%), and FAT (9.5%). Patients with FAT were the youngest at presentation (23.6 ± 18.5 years), followed by IART (28.8 ± 16.8 years), and atrial fibrillation (41.0 ± 17.2 years). They had more complex forms of heart disease (e.g., complex congenital heart disease in 56.5% compared with 35.2% in patients with atrial fibrillation) and more advanced NYHA functional class symptoms (e.g., 15.1% with NYHA functional class III or IV symptoms compared with 9.4% in patients with IART or atrial fibrillation).
Most patients with IART had moderate or complex congenital heart disease (85.9%). They had the largest systemic venous atrial diameters (52 ± 12 mm) and highest prevalence of pacemakers (41.7% compared with 30.3% in patients with FAT or atrial fibrillation).
Factors associated with atrial fibrillation
Factors associated with atrial fibrillation as the presenting arrhythmia are listed in Table 2. In multivariable analyses, older age (odds ratio [OR]: 1.024 per year; 95% confidence interval [CI]: 1.010 to 1.039; p = 0.001), hypertension (OR: 2.00; 95% CI: 1.08 to 3.71; p = 0.0290), and fewer cardiac surgeries (OR: 0.60; 95% CI: 0.49 to 0.75; p < 0.001) were independently associated with atrial fibrillation.
During a mean follow-up of 11.3 ± 9.4 years, an additional 5 (1.0%) patients developed FAT, 2 (0.4%) IART, and 42 (8.7%) atrial fibrillation. In univariable analyses, factors associated with new-onset atrial fibrillation were older age (hazard ratio [HR]: 1.039 per year; 95% CI: 1.018 to 1.059; p < 0.001) higher body mass index (HR: 1.057 per kg/m2; 95% CI: 1.009 to 1.107; p = 0.018), current smoking history (HR: 3.11; 95% CI: 1.30 to 7.46; p = 0.011), and atrial septal defect as type of congenital heart disease (HR: 1.98; 95% CI: 1.02 to 3.85; p = 0.044). In multivariable analysis, older age (HR: 1.033; 95% CI: 1.009 to 1.056; p = 0.0056) and atrial septal defect (HR: 2.26; 95% CI: 1.03 to 4.96; p = 0.041) remained independently associated with new-onset atrial fibrillation.
Demographics trends in type of atrial arrhythmias
As depicted in Figure 1, the proportion of arrhythmias due to IART increased with the complexity of congenital heart disease, from 47.2% to 62.1% to 67.0% in simple, moderate, and complex forms, respectively. By contrast, the proportion of arrhythmias due to atrial fibrillation decreased with disease complexity, from 44.9% to 30.1% to 21.6%, respectively (p = 0.0013). Variations in the proportion of arrhythmias due to FAT were less marked and ranged from 7.8% to 11.5%.
The type of atrial arrhythmias according to age is shown in Figure 2. IART was the predominant arrhythmia in younger patients, accounting for 75.0% of presenting arrhythmias in those <20 years of age. However, IART accounted for a decreasing proportion of arrhythmias as age increased (p < 0.0001), to a low of 44.2% in patients ≥50 years of age. Similarly, the proportion of presenting arrhythmias due to FAT decreased from 15.7% in patients <20 years to 4.7% in those ≥50 years of age. By contrast, atrial fibrillation accounted for 9.3% of arrhythmias in patients <20 years of age, and increased steadily with age to surpass IART as the most common presenting arrhythmia in patients ≥50 years of age (i.e., 51.2%). From a total of 136 patients who were at least 50 years of age by the end of follow-up, 73 (53.7%) had developed atrial fibrillation.
Pattern of atrial arrhythmias
Over the course of follow-up, the predominant pattern of atrial arrhythmias was paroxysmal in 62.3%, persistent in 28.2%, and permanent in 9.5%. However, the pattern varied according to type of presenting arrhythmia and age (Figure 3). The predominant pattern was paroxysmal for all arrhythmia types (Figure 3A). The second most common pattern was persistent for FAT and IART, and permanent for atrial fibrillation. A permanent pattern was observed in 20.7% of patients who presented with atrial fibrillation, in contrast to 4.6% of patients with IART and 6.7% with FAT (p < 0.0001).
Arrhythmia pattern according to age distribution is portrayed in Figure 3B. The predominant pattern was paroxysmal in all age categories except in those <20 years of age, in which persistent forms were observed in 50.4%. The proportion of paroxysmal and persistent arrhythmia patterns declined progressively from ages 20 to ≥50 years. By contrast, permanent arrhythmias increased with age, from 3.1% in patients <20 years of age to 5.5%, 12.9%, and 22.6% for ages 20 to 34 years, 35 to 49 years, and ≥50 years of age, respectively (p < 0.0001).
The main findings of this multicenter study assessing the types and patterns of atrial arrhythmias across a broad spectrum of patients with congenital heart disease include the following (Central Illustration): 1) IART is the most common atrial arrhythmia, followed by atrial fibrillation and FAT; 2) the prevalence of IART increases with complexity of congenital heart disease; 3) atrial fibrillation increases with age to surpass IART as the most common atrial arrhythmia in patients ≥50 years of age; 4) the predominant pattern of atrial arrhythmias is paroxysmal; and 5) atrial arrhythmias are increasingly long-lasting in older patients, with >20% of those ≥50 years of age exhibiting a permanent pattern.
In a population study based on administrative data, the estimated prevalence of atrial arrhythmias was 15.1% in adults with congenital heart disease, with projections surpassing 50% by the age of 65 years in those with complex defects (5). Atrial arrhythmias are associated with substantial morbidity and are the leading cause of hospitalizations in adults with congenital heart disease (3,4). They have been associated with syncope, heart failure, thromboembolic events, and sudden cardiac death (3,10,15,16). Common predisposing factors include intrinsic structural malformations, surgical scars, conduits, and patches, hypoxemia, and chronic pressure and/or volume overload, which, over time, result in arrhythmogenic remodeling changes with variable degrees of atrial dilation, hypertrophy, and fibrosis (3,17).
In patients with congenital heart disease, FATs predominantly arise in areas of abnormal conduction around suture points or scar borders. They are often nonautomatic and are thought to be due to micro–re-entry, owing to their capacity to be induced, terminated, and entrained by programmed atrial stimulation (18,19). FATs were the least common arrhythmias observed in the current study, and predominantly occurred in young patients (mean age 24 years) with moderate or complex defects. Macro–re-entrant IART circuits propagating around anatomic or electrical barriers to conduction have been well characterized in the congenital heart disease population. The observed association with greater complexity of congenital heart disease likely reflects extensive atrial scarring, such as from suture lines, baffles, and conduits. The high prevalence of pacemakers in patients with IART (i.e., 42%) is consistent with the coexistence of bradyarrhythmias (20). Loss of atrioventricular synchrony and longstanding sinus node dysfunction can result in ineffective atrial hemodynamics, which, in conjunction with scars and anatomic obstacles, establish a milieu for IART (3).
The increased long-term risk of developing atrial fibrillation in patients with atrial septal defects, regardless of the method of closure, has previously been described (21). The observed association between hypertension and atrial fibrillation is also well established in the general population (22,23). Elevated systemic blood pressure can promote atrial fibrillation through multiple mechanisms, including pulmonary venous atrial pressure overload and dilation, activation of the renin-angiotensin-aldosterone system, and induction of proinflammatory molecules, with accompanying electrical and structural remodeling effects (23–25). Similarly, the link between older age and atrial fibrillation is consistent across numerous studies (22,26), including one study on patients with tetralogy of Fallot (9). Age-related atrial arrhythmogenic changes include shortening of action potential duration, reduced capacity of the action potential to adapt to heart rate, and increased spatial variability of repolarization (27). However, the sharp increase in prevalence of atrial fibrillation appears to occur at a considerably younger age in patients with congenital heart disease when compared to the general population (22).
Considering the relatively young age of the first generation of long-term survivors with surgically palliated, severe congenital heart defects, it is not surprising that older patients have less complex disease. Identified age-related trends indicate that, as the growing population with congenital heart disease ages, atrial fibrillation supplants IART as the most common arrhythmia. In addition, as patients age, their atrial arrhythmias become increasingly refractory to rhythm-control treatment strategies. These observations underscore the potential for the arrhythmia burden to escalate, and for management issues to become increasingly challenging as atrial fibrillation develops in patients with complex congenital heart disease and arrhythmias become recalcitrant to therapy.
Our findings have important implications regarding anticipation of resource allocation and future research. For example, a better understanding of underlying mechanisms is required to improve preventive approaches and management strategies for atrial fibrillation. Whereas in the general population, the pathophysiology of atrial fibrillation centers on ion channel dysfunction, calcium handling abnormalities, structural remodeling, and autonomic neural dysregulation (22), it is unknown whether analogous mechanisms are at play in patients with congenital heart disease. Moreover, although randomized trials have demonstrated similar outcomes with rate-control and rhythm-control strategies in patients with atrial fibrillation, including those with heart failure (28), the merits of rate control have not been studied in the context of cyanotic heart disease, a systemic right ventricle, univentricular heart, or restrictive systemic or subpulmonary ventricular physiology. It remains to be determined whether preemptive strategies, such as surgical modifications, aggressive treatment of IART, and optimization of hemodynamics, could mitigate risks of developing atrial fibrillation.
The study is observational and hence subject to associated limitations, including the inability to adjust for unmeasured or unknown potential confounders. Although uniform definitions were applied by a blinded adjudicating committee in classifying arrhythmia types, decisions regarding therapy were at the discretion of treating physicians. This may have influenced factors such as designation of an atrial arrhythmia pattern as permanent, because it partly reflects a therapeutic approach, as opposed to an inherent pathophysiological attribute of the arrhythmia. Participating centers were encouraged to systematically include all eligible patients in order to limit selection bias. Nevertheless, presence of a documented atrial arrhythmia was an inclusion criterion, such that the study was not designed to calculate prevalence rates in the larger population with congenital heart disease. Rather, it was intended to shed light on the distribution and pattern of arrhythmias and age-related trends in a large well-characterized multicenter cohort of patients with arrhythmias and congenital heart disease. Generalizability of the findings is limited by the fact that all participating sites are referral centers for congenital heart disease, presumably enriching the study sample for more complex patients.
In this multicenter study that assessed atrial arrhythmia patterns and trends in a mixed cohort with congenital heart disease from across North America, IART was the most common type of presenting arrhythmia (>60% of cases), with significantly higher rates in patients with more complex disease. However, atrial fibrillation accounted for an increasing proportion of arrhythmias as patients aged, surpassing IART as the most common atrial arrhythmia in those ≥50 years of age. Although the predominant arrhythmia pattern was paroxysmal, arrhythmias became increasingly long-lasting with age and were considered permanent in nearly one-quarter of patients ≥50 years of age. Future research efforts centered on prevention and optimal management of arrhythmias in congenital heart disease should, therefore, consider these changing trends.
COMPETENCY IN MEDICAL KNOWLEDGE 1: IART is the most common atrial arrhythmia in patients with congenital heart disease, with a prevalence that increases according to complexity of the underlying cardiac defect.
COMPETENCY IN MEDICAL KNOWLEDGE 2: In patients with congenital heart disease, the prevalence of atrial fibrillation increases with age to surpass IART as the most common atrial arrhythmia in those 50 years of age and older.
COMPETENCY IN MEDICAL KNOWLEDGE 3: The predominant pattern of atrial arrhythmias in patients with congenital heart disease differs according to age, with a substantial proportion of older patients developing permanent arrhythmias.
TRANSLATIONAL OUTLOOK 1: It remains to be determined whether preventive strategies, such as prophylactic surgical approaches, prompt catheter ablation of IART, and judicious optimization of hemodynamics, could mitigate risks of developing atrial fibrillation and halt the progression toward permanent atrial arrhythmias.
TRANSLATIONAL OUTLOOK 2: Considering the age-related trends regarding the type and pattern of atrial arrhythmias in congenital heart disease, additional research is required to compare outcomes with rhythm-control versus rate-control treatment strategies.
The authors thank the following: Marie-Claude Villeneuve, MSc (Montreal Health Innovations Coordinating Center); Aynun Naher, MBBS, MS (Oregon Health and Science University); William R. Davidson Jr., MD, John J. Kelleman, MD, Elizabeth E. Adams, DO, and Dena Jefferson RN, BSN, CCRC (Hershey Medical Center); Morgan Hindes (Children's Hospital of Pittsburgh); Ryan Williams and Gwen Derk (University of California, Los Angeles); Michael G. Earing, MD, Jonathan W. Cramer, MD, and Emily Reinhardt, RN (Medical College of Wisconsin); and Meena Fatah (Hospital for Sick Children, Toronto).
Dr. Labombarda was supported by a grant from the Fédération Française de Cardiologie. Dr. Khairy was supported by a research chair in electrophysiology and congenital heart disease. The study was funded by an investigator-initiated unrestricted grant from Boehringer Ingelheim. The sponsor had no role in study design, data collection, analysis, interpretation, writing of the paper, and decision to submit the manuscript for publication. The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- Alliance for Adult Research in Congenital Cardiology
- confidence interval
- focal atrial tachycardia
- hazard ratio
- intra-atrial re-entrant tachycardia
- New York Heart Association
- odds ratio
- Received May 18, 2017.
- Revision received June 13, 2017.
- Accepted June 15, 2017.
- 2017 American College of Cardiology Foundation
- Marelli A.J.,
- Ionescu-Ittu R.,
- Mackie A.S.,
- Guo L.,
- Dendukuri N.,
- Kaouache M.
- Khairy P.,
- Ionescu-Ittu R.,
- Mackie A.S.,
- Abrahamowicz M.,
- Pilote L.,
- Marelli A.J.
- Khairy P.,
- Van Hare G.F.,
- Balaji S.,
- et al.
- Verheugt C.L.,
- Uiterwaal C.S.,
- van der Velde E.T.,
- et al.
- Bouchardy J.,
- Therrien J.,
- Pilote L.,
- et al.
- Khairy P.
- Gelatt M.,
- Hamilton R.M.,
- McCrindle B.W.,
- et al.
- Khairy P.,
- Aboulhosn J.,
- Gurvitz M.Z.,
- et al.,
- Alliance for Adult Research in Congenital Cardiology (AARCC)
- Khairy P.,
- Aboulhosn J.,
- Broberg C.S.,
- et al.,
- Anticoagulation Therapy in Congenital Heart Disease (TACTIC) investigators and the Alliance for Adult Research in Congenital Cardiology (AARCC)
- January C.T.,
- Wann L.S.,
- Alpert J.S.,
- et al.
- Warnes C.A.,
- Williams R.G.,
- Bashore T.M.,
- et al.
- Khairy P.
- Karunanithi Z.,
- Nyboe C.,
- Hjortdal V.E.
- Andrade J.,
- Khairy P.,
- Dobrev D.,
- Nattel S.
- Tremblay-Gravel M.,
- White M.,
- Roy D.,
- et al.
- McEwan P.E.,
- Gray G.A.,
- Sherry L.,
- Webb D.J.,
- Kenyon C.J.