Patients with adult congenital heart disease are at increased risk of ventricular arrhythmia (VA) and sudden cardiac death, although no clear predictors have been found. Ventricular programmed stimulation has been shown to predict clinical ventricular tachycardia and sudden death events, but the role of screening electrophysiology studies (S-EPSs) in this population remains poorly defined. Therefore, we sought to determine the prevalence of inducible VA and to evaluate the clinical predictors in a heterogeneous group of patients with adult congenital heart disease (≥18 years old) undergoing S-EPSs at preoperative or interventional cardiac catheterization. Studies for the primary evaluation of clinical VA were excluded. The demographic, clinical, and diagnostic findings were compared between the patients with positive and negative findings. From 2005 to 2009, 80 patients (mean age 30 ± 9 years) underwent S-EPSs, and 23 had inducible VA. The diagnoses for those with studies positive for VA included tetralogy of Fallot (n = 12), d-transposition of the great arteries (n = 6), pulmonary stenosis (n = 2), double outlet right ventricle (n = 1), double inlet left ventricle (n = 1), and Ebstein’s anomaly (n = 1). Men were significantly more likely to have a S-EPS positive for VA (p = 0.015). Increasing QRS duration, decreasing peak oxygen uptake (percentage of predicted), and ventricular fibrosis with cardiovascular magnetic resonance imaging were significantly associated with studies positive for VA (p <0.05). Combined fibrosis and a peak oxygen uptake <80% of predicted had 100% sensitivity for positive VA findings. In conclusion, almost 30% of those with adult congenital heart disease undergoing S-EPSs had inducible VA. A prolonged QRS duration, diminished exercise capacity, and the presence of ventricular fibrosis were significantly associated with findings positive for VA and might improve patient selection for screening evaluations.
Ventricular arrhythmias (VAs) associated with sudden death are major contributors to late mortality in patients with repaired tetralogy of Fallot (TOF) and might be more common in others with adult congenital heart disease (ACHD) than previously thought. Electrophysiologic testing with programmed ventricular stimulation has been shown to have prognostic value in risk stratifying clinical ventricular tachycardia (VT) and sudden cardiac death in patients with TOF. However, the role of screening electrophysiology studies (S-EPSs) in the general ACHD population remains poorly defined, particularly in the absence of clinical arrhythmia. Therefore, we sought to determine the prevalence of inducible VA in a heterogeneous group of ACHD undergoing S-EPSs at cardiac catheterization. Our secondary objective was to evaluate the association between previously identified markers of sudden death and inducible VAs in a mixed ACHD cohort.
Methods
We retrospectively evaluated 80 patients with ACHD (≥18 years old) who had undergone S-EPSs at Nationwide Children’s Hospital from 2005 to 2009. A S-EPS was defined as programmed ventricular stimulation performed routinely at cardiac catheterization for preoperative evaluation or planned transcatheter intervention but excluded electrophysiologic testing when performed primarily for evaluation of documented clinical VT. The patients included those with TOF, d-transposition of the great arteries, and other forms of congenital heart disease with presumed ventricular hemodynamic impairment. Our hospital’s institutional review board approved the study.
The S-EPS was performed using standard pacing protocols. Programmed stimulation was performed at 2 sites (ventricular apex and outflow tract) with 8-beat drive trains (cycle lengths of 400 to 600 ms) and ≤3 extra-stimuli with coupling intervals of ≥180 ms. In the absence of inducible VA, the protocol was repeated with isoproterenol infusion to increase the heart rate by ≥15%. Antiarrhythmic medications were withheld for ≥5 half-lives before the study, including sotalol in 6 patients with atrial arrhythmia and mexilitine in 1 patient with symptomatic premature ventricular beats. Amiodarone was used by 2 patients for atrial arrhythmias and was not withheld because of its long half-life.
The primary end point of inducible VA was a composite of nonsustained VT (≥10 beats), sustained VT (persisting ≥30 seconds), and hemodynamically unstable VT (monomorphic and polymorphic), and ventricular fibrillation. The S-EPS findings were classified as either inducible (+VA) or noninducible (−VA) for VA.
Details of the demographic variables, anthropometrics, congenital heart disease diagnoses, surgical history, cardiovascular symptoms, New York Heart Association functional class, cardiovascular medications, and findings from electrocardiography, 24-hour Holter monitoring, cardiopulmonary exercise testing, cardiovascular magnetic resonance imaging, computed tomography, cardiac catheterization, and S-EPSs were collected.
The surgical characteristics included a history of palliative shunt and the presence of a ventriculotomy incision or transannular patch. Electrocardiographic data included the presence of a right bundle branch block, QRS duration, and corrected QT intervals. Ventricular-paced rhythms were excluded from the electrocardiographic analysis. Atrial and ventricular ectopy were captured from the Holter monitors, and ventricular premature complexes and nonsustained VT were risk stratified using the modified Lown criteria. Cardiopulmonary exercise testing data included the peak oxygen uptake (ml O 2 /kg/min and percentage of predicted), and associated ventricular ectopy. Cardiovascular magnetic resonance imaging and computed tomographic data included the right and left ventricular dimensions and systolic function. Late gadolinium enhancement suggestive of ventricular fibrosis was recorded. The cardiac catheterization data included the right and left ventricular systolic and end-diastolic pressures. Pulmonary capillary wedge pressure was used if the systemic ventricular end-diastolic pressure was not measured.
Standard definitions were used for the sensitivity, specificity, and predictive values of the clinical markers in identifying patients with +VA. Continuous variables are presented as the mean ± SD. Logistic regression analysis was used to identify significant risk factors associated with +VA. Fisher’s exact test was used to evaluate for significant differences in the congenital heart disease diagnoses and +VA. Two-tailed values of p <0.05 were considered statistically significant.
Results
A total of 50 studies were performed during preoperative cardiac catheterization for hemodynamic evaluation, and 30 studies were performed at transcatheter intervention. Patients were generally asymptomatic, with 64% having New York Heart Association functional class I. Congestive heart failure medications were used by 28 patients (35%).
The congenital heart disease diagnoses included 37 patients with repaired TOF (46%), 22 with d-transposition of the great arteries status after the Mustard or Senning procedure (28%), 13 with pulmonary stenosis status after valvotomy (16%), and 2 with double-outlet right ventricle after a Rastelli-type repair (3%). The remaining 6 patients had single ventricle anatomy, including 2 with a double-outlet right ventricle, 2 with a double-inlet left ventricle, 1 with Ebstein’s anomaly of the tricuspid valve, and 1 with tricuspid atresia. Overall, 23 patients (29%) had undergone initial palliative shunting, and a ventriculotomy incision or transannular patch had been used in 45 repairs (56%). The mean age at complete repair was 5 ± 6 years, and the mean interval from surgical correction to the S-EPSs was 24 ± 9 years.
+VA was found in 23 patients undergoing S-EPS (29%), with 87% induced at baseline without the isoproterenol infusion. Nonsustained VT was induced in 10 studies, including 6 monomorphic and 4 polymorphic VTs ( Figure 1 ). One patient had sustained VT that self-terminated. Hemodynamically unstable +VA studies included monomorphic VT (n = 5), polymorphic VT (n = 2), and ventricular fibrillation (n = 5).
Of the patients with a +VA study, 14 were found before operative repair, including pulmonary valve replacement for severe pulmonary insufficiency (n = 11), right ventricle-to-pulmonary artery conduit replacement (n = 1), intra-atrial baffle repair (n = 1), and Fontan completion (n = 1). +VA was found during 7 transcatheter interventions, including stent placement for superior vena cava baffle stenosis (n = 4), baffle leak closure (n = 1), pulmonary arterioplasty and stent placement (n = 1), and coil embolization of aortopulmonary collateral arteries (n = 1). +VA was also found in 2 patients in whom no transcatheter intervention was necessary.
The baseline characteristics of the +VA and −VA groups are listed in Table 1 . No significant difference was found in +VA across the ACHD diagnoses, although the small group sizes might have been insufficient to detect any differences. No significant associations were found between +VA and previous palliative shunt, ventriculotomy incision, age at surgical repair, age at the S-EPS, cardiovascular symptoms, or cardiovascular medication use. Both patients taking amiodarone had −VA, and 2 patients receiving chronic sotalol therapy had +VA. Men were significantly more likely to have +VA (p = 0.015).
| Variable | +VA (n = 23) | −VA (n = 57) | Odds Ratio | 95% CI | p Value |
|---|---|---|---|---|---|
| Congenital heart disease diagnosis | 0.697 ⁎ | ||||
| Tetralogy of Fallot | 12 (52%) | 25 (44%) | |||
| d-Transposition of the great arteries | 6 (26%) | 16 (28%) | |||
| Pulmonary stenosis | 2 (9%) | 11 (19%) | |||
| Double-outlet right ventricle | 1 (4%) | 1 (2%) | |||
| Single ventricle | 2 (9%) | 4 (7%) | |||
| Height (m) | 1.7 ± 0.1 | 1.7 ± 0.2 | 1.04 | 0.98–1.09 | 0.193 |
| Weight (kg) | 74 ± 14 | 73 ± 21 | 1.00 | 0.98–1.02 | 0.863 |
| Women | 5 (22%) | 30 (53%) | 0.25 | 0.08–0.77 | 0.015 |
| Previous palliative shunt | 4 (17%) | 19 (33%) | 0.42 | 0.13–1.41 | 0.161 |
| Ventriculotomy | 14 (61%) | 31 (54%) | 1.3 | 0.49–3.5 | 0.597 |
| Age at complete repair (years) | 6 ± 6 | 5 ± 6 | 1.01 | 0.94–1.09 | 0.728 |
| Age at electrophysiology study (years) | 28 ± 7 | 31 ± 9 | 0.96 | 0.9–1.02 | 0.201 |
| Symptoms | |||||
| Dyspnea | 8 (35%) | 18 (32%) | 1.16 | 0.42–3.21 | 0.782 |
| Palpitations | 7 (30%) | 13 (23%) | 1.48 | 0.5–4.37 | 0.477 |
| Fatigue | 8 (35%) | 11 (19%) | 2.23 | 0.76–6.58 | 0.146 |
| Angina pectoris | 2 (9%) | 11 (19%) | 0.4 | 0.08–1.96 | 0.257 |
| Syncope | 2 (9%) | 5 (9%) | 0.99 | 0.18–5.51 | 0.991 |
| New York Heart Association functional class II or greater | 10 (43%) | 19 (33%) | 1.54 | 0.57–4.15 | 0.394 |
| Cardiovascular medications | |||||
| Angiotensin-converting enzyme inhibitor | 6 (26%) | 13 (23%) | 1.19 | 0.39–3.65 | 0.755 |
| Diuretic | 4 (17%) | 7 (12%) | 1.5 | 0.39–5.72 | 0.55 |
| Digoxin | 3 (13%) | 7 (12%) | 1.07 | 0.25–4.56 | 0.926 |
| β Blocker | 4 (17%) | 5 (9%) | 2.19 | 0.53–9.02 | 0.278 |
| Antiarrhythmic drug | 2 (9%) | 7 (12%) | 0.68 | 0.13–3.55 | 0.648 |
A history of syncope was found in 7 patients (2 with +VA and 5 with −VA), including 6 with d-transposition of the great arteries and 1 with pulmonary stenosis. Of those with +VA, 1 was diagnosed with narcolepsy, and 1 was diagnosed with heat exhaustion. Of the patients with −VA, 2 had had remote syncope episodes (>1 year before EPS) and 2 had probable vasodepressor syncope. The remaining patient underwent cardiac catheterization for hemodynamic assessment of possible intra-atrial baffle obstruction after recurrent episodes of syncope. Although he had a −VA study, he was found to have inducible atrial arrhythmias and β-blockade therapy was started.
Electrocardiograms without ventricular pacing were available for 90% of the patients, Holter monitoring results for 93%, cardiopulmonary exercise testing findings for 65%; cardiac magnetic resonance imaging findings for 83%, and cardiac computed tomographic findings for 10%.
Of the 80 patients, 68 had a baseline sinus rhythm and 4 a junctional escape rhythm. +VA was significantly associated with right bundle branch block (p = 0.024) and increasing QRS duration (p = 0.012). However, the corrected QT interval was not significantly associated with +VA ( Table 2 ).
| Variable | +VA (n = 19) | −VA (n = 53) | OR | 95% CI | p Value |
|---|---|---|---|---|---|
| Right bundle branch block | 17 (90%) | 31 (58%) | 6.03 | 1.26–28.81 | 0.024 |
| QRS duration (ms) | 154 ± 24 | 132 ± 30 | 1.07 | 1.01–1.05 | 0.012 |
| QRS duration >180 ms | 3 (16%) | 4 (8%) | 2.25 | 0.45–11.15 | 0.321 |
| Corrected QT interval (ms) | 462 ± 28 | 447 ± 33 | 1.02 | 0.998–1.03 | 0.074 |
Holter monitoring demonstrated premature atrial beats in 76% of the patients and premature ventricular beats in 91%. Atrial tachycardia was recorded in 14 patients (19%). High-risk ventricular ectopy was found in 39% using the modified Lown criteria (≥2 consecutive premature ventricular contractions). In the entire cohort, 14% had a history of nonsustained VT. No significant association was found between +VA and either atrial or ventricular arrhythmia found on ambulatory monitoring or any history of nonsustained VT.
Overall, the patient exercise capacity was diminished, with a mean peak oxygen uptake of 69% of predicted. No significant association was found between +VA and the absolute peak oxygen uptake. However, after adjusting for gender, age, height, and weight, a significant association was found between +VA and a decrease in the percentage of the predicted maximum oxygen uptake (p = 0.026; Table 3 ).
| Variable | +VA (n = 17) | −VA (n = 35) | OR | 95% CI | p Value |
|---|---|---|---|---|---|
| Peak oxygen uptake (ml O 2 /min/kg) | 25 ± 8 | 28 ± 12 | 0.96 | 0.9–1.04 | 0.338 |
| Peak oxygen uptake (percentage of predicted ⁎ ) | 61 ± 15 | 73 ± 18 | 1.05 | 1.01–1.09 | 0.026 |
| Exercise-associated ventricular ectopy | 5 (29%) | 7 (20%) | 2.18 | 0.6–7.96 | 0.237 |
Ventricular dimensions and function were a composite of the cardiovascular magnetic resonance imaging and computed tomographic measurements, and the results were grouped according to the systemic or subpulmonary ventricle ( Table 4 ). No significant association between +VA and ventricular volumes or systolic function were found. Late gadolinium enhancement was performed in 45 studies. Ventricular fibrosis was found in 42% and was significantly associated with +VA (p = 0.002). The sensitivity of fibrosis for +VA was 92%, and the specificity was 76%. The negative predictive value was 96% ( Figure 2 ).
| Variable | +VA (n = 21) | −VA (n = 56) | OR | 95% CI | p Value |
|---|---|---|---|---|---|
| Systemic ventricle | |||||
| End-diastolic volume (ml/m 2 ) | 74 ± 27 | 68 ± 28 | 1.01 | 0.99–1.03 | 0.36 |
| End-systolic volume (ml/m 2 ) | 40 ± 19 | 35 ± 19 | 1.01 | 0.98–1.04 | 0.29 |
| Ejection fraction (%) | 47 ± 9 | 51 ± 10 | 0.96 | 0.91–1.01 | 0.15 |
| Subpulmonary ventricle | |||||
| End-diastolic volume (ml/m 2 ) | 100 ± 40 | 99 ± 40 | 1.0 | 0.99–1.01 | 0.90 |
| End-systolic volume (ml/m 2 ) | 59 ± 30 | 54 ± 25 | 1.01 | 0.99–1.03 | 0.52 |
| Ejection fraction (%) | 44 ± 11 | 46 ± 11 | 0.98 | 0.93–1.03 | 0.39 |
| Ventricular fibrosis | 11/12 (92%) | 8/33 (24%) | 34.4 | 3.8–309.2 | 0.002 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
