Atrial arrhythmias are associated with an increased mortality risk in adults with congenital heart disease (CHD). However, little is known about risk stratification in the specific group of adult patients with CHD and atrial arrhythmias. We sought to identify predictors of mortality in adult with CHD and atrial arrhythmias and to establish a risk score. The study involved 378 adult patients with CHD (mean age 39 ± 13 years) and atrial arrhythmias who had serial follow-up in a tertiary referral center from 1999 through 2009. During a median follow-up of 5.2 years, there were 40 deaths (11%). Overall mortality rate was 2.0% per patient-year. Common modes of death included heart failure–related death (35%), sudden cardiac death (20%), and perioperative death (18%). Independent predictors of mortality were poor functional class (hazard ratio 3.69, 95% confidence interval [CI] 1.69 to 8.03, p = 0.001), single-ventricle physiology (hazard ratio 3.33, 95% CI 1.51 to 7.35, p = 0.003), pulmonary hypertension (hazard ratio 2.96, 95% CI 1.41 to 6.19, p = 0.004), and valvular heart disease (hazard ratio 2.73, 95% CI 1.33 to 5.59, p = 0.006). A risk score was constructed using these predictors in which patients were assigned 1 point for the presence of each risk factor. Mortality rates in the low-risk (no risk factor), moderate-risk (1 risk factor), and high-risk (>1 risk factor) groups were 0.5%, 1.9%, and 6.5% per patient-year, respectively (log-rank p <0.001). In conclusion, in adult with CHD and atrial arrhythmias specific clinical variables identify patients at high risk for death. Importantly, the absence of any of these risk factors is associated with an excellent survival despite the presence of atrial arrhythmias.
Atrial arrhythmias are common late sequelae (∼15%) in adult patients with congenital heart disease (CHD). As demonstrated by 2 recent large adult CHD population-based studies, atrial arrhythmias are associated with an approximately twofold increased risk of mortality. However, the adult CHD population is a very heterogenous group with different background risks for death. There is a lack of data on risk stratification in the specific population of adult patients with CHD and atrial arrhythmias. It is important to be able to identify a patient at high absolute risk for death because this will affect patient surveillance and management. We sought to identify predictors of mortality in adult patients with CHD and atrial arrhythmias and to establish a risk score.
Methods
All adult patients with CHD and atrial arrhythmia who were followed serially from January 1999 through December 2009 in the Toronto Congenital Cardiac Centre for Adults were included. We chose these dates to minimize missing data and provide contemporary data. Patients were identified by a search of the institutional database. The study group included all patients with a clinical presentation of documented sustained intra-atrial reentrant tachycardia or atrial fibrillation. Patients who had nonsustained episodes of atrial arrhythmia during Holter monitoring, atrioventricular nodal reentry tachycardia, or atrioventricular reentry tachycardia were not included. The study was approved by the institutional ethics review board.
Clinical information from the institutional database was supplemented by a detailed retrospective review of medical records and echocardiographic and cardiac catheterization data. Baseline characteristics included demographics, medical history (i.e., CHD diagnosis, cardiac interventions, hypertension requiring medical treatment, diabetes mellitus requiring medical treatment, stroke, transient ischemic attack, and systemic embolism), New York Heart Association (NYHA) functional class, and cardiac medications. Functional status was determined at the time of the index visit using the NYHA classification. Complexity of CHD was classified as simple, moderate, or complex according to previously published guidelines.
Echocardiographic data were obtained from reports at the time of the index visit (first visit during study period) and included ventricular systolic dysfunction, valvular heart disease, intracardiac shunt, and systolic pulmonary arterial pressure. Mean pulmonary arterial pressure was obtained from cardiac catheterization data when available.
For the univariate and multivariate models, significant subaortic ventricular systolic dysfunction was defined as moderate or severe subaortic ventricular systolic dysfunction. Subaortic ventricular systolic function was assessed echocardiographically in all patients. Subaortic ventricular systolic function was graded by visual assessment and classified as normal (ejection fraction ≥50%), mildly (ejection fraction 40% to 49%), moderately (ejection fraction 30% to 39%), or severely (ejection fraction <30%) hypokinetic. Valvular heart disease was defined as moderate or severe stenosis or regurgitation according to current guidelines. Pulmonary hypertension was defined as mean pulmonary arterial pressure at rest >25 mm Hg by cardiac catheterization or systolic pulmonary arterial pressure >40 mm Hg by echocardiography.
Cause of death was classified as sudden death, death secondary to heart failure, perioperative death, other cardiovascular death, and noncardiovascular death. Sudden death was defined as death occurring within 1 hour of onset of symptoms or nonwitnessed death during sleep. Death secondary to progressive heart failure was defined as death owing to progressive myocardial failure of the subaortic or subpulmonic ventricle. Death was considered perioperative if it occurred before hospital discharge or within 30 days of cardiac surgery. Other causes of cardiovascular death included myocardial infarction, pulmonary emboli, occlusion of shunt, severe hemoptysis, stroke, rupture of aneurysm, and endocarditis. Other causes of death were classified as noncardiovascular death. Deaths were classified as “unknown” if the cause of death could not be accurately determined.
Statistical analyses were performed with SPSS 15.0 (SPSS, Inc., Chicago, Illinois). Continuous data are presented as mean ± SD or median (range) as appropriate. Categorical variables are represented by frequencies and percentages. Time 0 (index visit) was defined as the time of the first patient visit during the study period. Patient-years were accrued from time of entry until occurrence of death, study termination, or loss to follow-up. Overall survival was plotted using the Kaplan–Meier method and log-rank statistics were used to compare risk categories.
To determine predictors of mortality, univariate and multivariate Cox proportional hazard models were constructed after proportional hazard assumptions were verified. The following variables were entered into the multivariate Cox model (forward stepwise): age, male gender, prosthetic heart valve, subaortic ventricular dysfunction, previous stroke/transient ischemic attack/systemic embolism, single-ventricle physiology, valvular heart disease, pulmonary hypertension, and NYHA functional class ≥III. Relative values of beta coefficients calculated from the multivariate model were used to devise a risk score to estimate overall mortality. Two-tailed probability values <0.05 were considered statistically significant.
Results
In total 378 adult patients with CHD and sustained atrial arrhythmias were included. Baseline characteristics including types of arrhythmias are presented in Table 1 . Of all study patients 212 patients (56%) had a history of emergency room visit or hospital admission for sustained atrial arrhythmia. Two hundred fifty-one patients (66%) had paroxysmal atrial arrhythmias; the remainder (n = 127, 34%) had persistent/permanent atrial arrhythmias. Of patients with paroxysmal atrial arrhythmia most (n = 236, 94%) had a clinical presentation requiring intervention including antiarrhythmic therapy in 220, anticoagulation in 146, cardioversion in 102, and catheter ablation in 24.
| Age (years) | 39 ± 13 |
| Men | 199 (53%) |
| Complex congenital heart disease | 162 (43%) |
| Intra-atrial reentrant tachycardia | 239 (63%) |
| Atrial fibrillation | 128 (34%) |
| Atrial fibrillation and intra-atrial reentrant tachycardia | 11 (3%) |
| Arterial hypertension | 26 (7%) |
| Diabetes mellitus | 9 (2%) |
| Previous stroke, transient ischemic attack, or systemic embolism | 40 (11%) |
| Previous stroke | 25 |
| Previous transient ischemic attack | 13 |
| Previous systemic embolism | 2 |
| Prosthetic heart valve | 105 (28%) |
| Bioprosthesis/homograft | 60 |
| Mechanical prosthesis | 45 |
| Pacemaker | 73 (19%) |
| Implantable cardioverter–defibrillator | 9 (2%) |
| New York Heart Association functional class | |
| I | 240 (64%) |
| II | 107 (28%) |
| III | 31 (8%) |
| Medications | |
| warfarin | 236 (62%) |
| Aspirin | 48 (13%) |
| Angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker | 130 (34%) |
| Loop diuretics | 92 (24%) |
| Spironolactone | 24 (6%) |
| Antiarrhythmic drugs | 285 (75%) |
| Digoxin | 135 |
| Amiodarone | 106 |
| β Blockers | 104 |
| Sotalol | 47 |
| Calcium channel blockers | 18 |
| Class 1 antiarrhythmic drugs | 7 |
| Valvular heart disease ⁎ | 153 (41%) |
| Venous atrioventricular regurgitation | 60 |
| Subaortic atrioventricular regurgitation | 52 |
| Pulmonary regurgitation | 41 |
| Subaortic atrioventricular stenosis | 18 |
| Aortic stenosis | 16 |
| Pulmonary stenosis | 12 |
| Aortic regurgitation | 11 |
| Venous atrioventricular stenosis | 1 |
| Ventricular systolic dysfunction † | 151 (40%) |
| Subaortic ventricular systolic dysfunction | 103 |
| Subpulmonic ventricular systolic dysfunction | 63 |
| Intracardiac shunt | 89 (24%) |
| Pulmonary hypertension ‡ | 80 (21%) |
⁎ At least moderate stenosis or regurgitation.
† At least moderate subaortic or subpulmonic ventricular systolic dysfunction.
‡ Mean pulmonary arterial pressure at rest >25 mm Hg by cardiac catheterization or systolic pulmonary arterial pressure >40 mm Hg by echocardiography.
A large proportion of patients were classified as having complex CHD (43%; Tables 1 and 2 ). Cyanosis was present in 18 of 75 patients (24%) with single-ventricle physiology primarily restricted to those with a palliative shunt (92% of patients with palliative shunts were cyanotic). Most patients with pulmonary hypertension had left-sided valvular disease (39%) or a diagnosis of (repaired) intracardiac shunt (45%). Common valvular heart disease included atrioventricular valve regurgitation or pulmonary regurgitation particularly in patients with tetralogy of Fallot.
| Single-ventricle physiology | 75 (20%) |
| Atriopulmonary connection | 39 |
| Palliative shunt | 13 |
| Right atrial–right ventricular connection | 12 |
| Intra-atrial lateral tunnel | 9 |
| Extracardiac conduit | 2 |
| Tetralogy of Fallot, repaired | 58 (15%) |
| Complete transposition of great arteries | 53 (14%) |
| Atrial switch procedure | 46 |
| Rastelli repair | 5 |
| Arterial switch procedure | 2 |
| Atrial septal defect | 40 (11%) |
| Repaired | 30 |
| Unrepaired | 10 |
| Atrioventricular septal defect, repaired | 25 (7%) |
| Congenital corrected transposition of great arteries | 20 (5%) |
| Conventional repair | 18 |
| Double switch procedure | 2 |
| Ebstein malformation of tricuspid valve | 18 (5%) |
| Repair/replacement of tricuspid valve | 8 |
| No tricuspid valve surgery | 10 |
| Ventricular septal defect | 15 (4%) |
| Repaired | 10 |
| Unrepaired | 5 |
| Other congenital heart disease ⁎ | 74 (20%) |
⁎ These include congenital aortic valve disease, partial/total anomalous pulmonary venous connection, congenital mitral valve disease, congenital pulmonary valve disease, aortic coarctation, double-outlet right ventricle, bicuspid aortic valve, patent foramen ovale, Eisenmenger syndrome, and truncus arteriosus.
During a median follow-up of 5.2 years (range 0.3 to 10.6) 40 deaths (11%) occurred. Characteristics of mode of death are presented in Table 3 . Primary cardiovascular causes of death were death secondary to progressive heart failure (35%), sudden death (20%), perioperative death (18%), and other cardiovascular causes (5%). Noncardiovascular causes occurred in 4 patients (10%). In 5 patients (13%) cause of death was unknown. Crude mortality rate was 2.0% (95% confidence interval [CI] 1.4 to 2.7) per person-year. Cumulative survival rates at 2, 5, and 8 years were 97.1 ± 0.9%, 91.3 ± 1.7%, and 84.8 ± 2.6%, respectively ( Figure 1 ).
| Mode of death | Number of Patients | Age at Death, Median (range) | Ventricular Dysfunction ⁎ | CHD Diagnosis (Patients) | Details of Death (Patients) |
|---|---|---|---|---|---|
| Heart failure | 14 (35%) | 44 (29–69) | 14 (100%) | SV (4), TGA (3), TOF (2), Ebstein anomaly (1), cc-TGA (1), AVSD (1), PAPVC (1), MR (1) | contraindication for heart transplantation (7), no formal heart transplantation assessment (4), waiting for heart transplant (2), declined heart transplantation (1) |
| Sudden death | 8 (20%) | 35 (21–48) | 4 (50%) | SV (3), TGA (2), AVSD (2), AR (1) | arrhythmia/sudden collapse (6), no witnessed death during sleep (2) |
| Perioperative | 7 (18%) | 44 (34–58) | 5 (71%) | SV (3), TOF (1), cc-TGA (1), TGA (1), Eisenmenger syndrome (1) | Fontan conversion (3), heart transplantation (2), pulmonary arterial band for arterial switch conversion (1), conduit revision (1) |
| Cardiovascular | 2 (5%) | 40 (37–43) | 1 (50%) | aortic coarctation (1), cc-TGA (1) | aortic aneurysmal rupture (1), hemoptysis in presence of aortopulmonary collaterals (1) |
| Noncardiovascular | 4 (10%) | 40 (30–60) | 1 (25%) | SV (1), ASD (1), AVSD (1), Shone complex (1) | septic shock (2), severe COPD (1), hemorrhage after pelvic surgery for twisted ovarian cysts (1) |
| Unknown | 5 (13%) | 51 (45–63) | 2 (40%) | TOF (1), AS (1), PS (1), cc-TGA (1), AR (1) | no details on death |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
