OUP user menu

Atrial-based pacing has no benefit over ventricular pacing in preventing atrial arrhythmias in adults with congenital heart disease

Petra Opić, Sing-Chien Yap, Matthijs Van Kranenburg, Arie P. Van Dijk, Werner Budts, Hubert W. Vliegen, Lieselot Van Erven, Anil Can, Gulhan Sahin, Natasja MS. De Groot, Maarten Witsenburg, Jolien W. Roos-Hesselink
DOI: http://dx.doi.org/10.1093/europace/eut213 1757-1762 First published online: 12 July 2013


Aims To determine whether atrial-based pacing prevents atrial arrhythmias in adults with congenital heart disease (CHD) compared with ventricular pacing.

Methods and results All adult CHD patients from four participating centres with a permanent pacemaker were identified. Patients with permanent atrial arrhythmias at pacemaker implantation and patients who received a pacemaker for treatment of drug-refractory atrial arrhythmias were excluded. The final study population consisted of 211 patients (52% male, 36% complex CHD) who received a first pacemaker for sick sinus dysfunction (n = 82) or atrioventricular block (n = 129) at a median age of 24 years [interquartile range (IQR), 12–34]. A history of atrial arrhythmias at implantation was present in 49 patients (23%). Atrial-based pacing was the initial pacing mode in 139 patients (66%) while the others (34%) received ventricular pacing. During a median follow-up of 13 years (IQR, 7–21), 90 patients (43%) developed an atrial arrhythmia. Multivariate analysis demonstrated no significant effect of atrial-based pacing on subsequent atrial arrhythmias [hazard ratio (HR), 1.53; 95% confidence interval (CI), 0.91–2.56; P = 0.1]. Independent predictors of atrial arrhythmia were history of atrial arrhythmias (HR, 5.55; 95% CI, 3.47–8.89; P< 0.0001), older age (≥18 years) at pacemaker implantation (HR, 2.29; 95% CI, 1.29–4.04; P = 0.005), and complex CHD (HR, 1.57; 95% CI, 1.01–2.45; P = 0.04).

Conclusion In contrast to the general population, atrial-based pacing was not associated with a lower incidence of atrial arrhythmia in adults with CHD.

  • Pacing
  • Tetralogy of Fallot
  • Transposition of the Great Vessels
  • Arrhythmia
  • Atrial fibrillation

What's new?

  • There is a high incidence of atrial arrhythmias in adults with congenital heart disease who receive permanent pacemaker therapy.

  • Atrial-based pacing does not seem to reduce the incidence of atrial arrhythmias compared with ventricular pacing in this specific population.

  • A history of atrial arrhythmias, older age, and complex congenital heart disease are independent predictors of atrial arrhythmias in this specific population.


Atrial arrhythmias play an important role in the management of adults with congenital heart disease (CHD).1 The prevalence of atrial arrhythmias is ∼15% in the adult CHD population, and the risk of developing atrial arrhythmias increases steadily with age.2 Atrial arrhythmias contribute significantly to morbidity and mortality in this population.24 Therefore, strategies to reduce the burden of atrial arrhythmias in this population are highly welcome.

An interesting subgroup is the adult CHD population with bradycardia indications for permanent pacemaker implantation. In the general population, several prospective randomized trials have clearly demonstrated that atrial-based pacing (‘physiological pacing’) reduces the incidence of atrial fibrillation compared with ventricular pacing.510 This is not surprising, as loss of atrioventricular (AV) synchrony due to ventricular pacing is associated with increased natriuretic peptide levels, sympathetic nervous activation, and atrial pressures; all factors associated with the development of atrial fibrillation.11,12 Whether atrial-based pacing is also beneficial in the adult CHD population is not clear.13 The aim of the present multicentre study is to investigate whether atrial-based pacing prevents atrial arrhythmias in adults with CHD.


For the present retrospective observational study, all adults with CHD and pacemaker implantation were identified using the nationwide CONgenital CORvita (CONCOR) registry in the Netherlands, and a Belgian tertiary care adult CHD database.14 The central medical ethics committee in the Netherlands and the local Belgian Ethics Board approved the protocol. This study complies with the declaration of Helsinki.

Data were collected from medical records and pacemaker databases. Baseline data prior to pacemaker implantation were registered from the patient records including gender, congenital anatomic diagnosis, surgical procedures, and history of atrial arrhythmias. The complexity of the CHD diagnosis was defined as simple, moderate, or complex according to the classification adopted at the American Heart Association Task Force on Adults with CHD.15 Detailed information concerning pacemaker implantation was recorded and included age at implantation, indication for pacemaker implantation, method of implant (endocardial/epicardial), pacing mode (AAI/VVI/DDD/VDD), and the occurrence of sustained atrial arrhythmias during follow-up. Atrial-based pacing, which includes AAI, DDD, and VDD, was compared with ventricular pacing (VVI). The endpoint was the occurrence of sustained atrial arrhythmias (i.e. intra-atrial reentrant tachycardia, atrial tachycardia, and atrial fibrillation) documented by electrocardiogram, Holter, or stored logs.

Statistical analysis

Continuous data are presented as mean ± SD or median with interquartile ranges (IQRs) as appropriate. Categorical variables are represented by frequencies and percentages. Comparison of continuous variables between groups was made by unpaired Student's t-tests. In the case of a skewed distribution, the Mann–Whitney U test was used. When comparing frequencies, the χ2 test or Fisher's exact test was used, where appropriate.

Cumulative freedom from atrial arrhythmias was constructed with the use of the Kaplan–Meier method and groups were compared by log-rank statistics. To determine predictors of sustained atrial arrhythmias during follow-up, univariate and multivariate Cox proportional hazard models were constructed after proportional hazard assumptions were verified. The patient was set as unit of analysis and time to first atrial arrhythmia was determined. Candidate variables included: atrial-based pacing, older age (≥18 years) at pacemaker implantation, complex CHD, female gender, history of atrial arrhythmia, endocardial pacing system, and sick sinus syndrome. Two-tailed probability values of <0.05 were considered statistically significant. Statistical analysis was performed using the statistical package R (64 bit) for Mac, version 2.14.2.


Patient characteristics

Crosscheck with local pacemaker registries of the four participating tertiary centres revealed a total of 274 patients. After exclusion of 63 patients with permanent or drug-refractory atrial arrhythmias at implantation, 211 patients comprised the final study population. The baseline characteristics of the included patients are summarized in Table 1. Patients with complex CHD comprised 37% of the population, and were mainly patients with transposition of the great arteries and single ventricle physiology. The majority of patients (66%) received atrial-based pacing at initial implantation (Figure 1). There were important baseline differences between patients with atrial-based pacing and those with ventricular pacing. Patients with ventricular pacing more often had the first pacemaker implantation at paediatric/adolescent age (64 vs. 22%, P< 0.001) and in an older era (before 1994) (75 vs. 21%, P< 0.001). Furthermore, an epicardial pacing system was more often used in patients with ventricular pacing (49 vs. 25%, P< 0.001). There was no difference in the complexity of the underlying CHD or the prevalence of a history of atrial arrhythmias between the two groups.

View this table:
Table 1

Baseline characteristics at first pacemaker implantation

VariableTotal (n = 211)Atrial-based pacing (n = 139)Ventricular pacing (n = 72)P value
Age at first implantation (years)a24 (12–34)26 (19–39)12 (7–26)<0.001
First implant <18 years77 (36%)31 (22%)46 (64%)<0.001
Male gender110 (52%)72 (52%)38 (53%)0.9
Complex CHD79 (37%)52 (37%)27 (38%)1.0
History of atrial arrhythmia49 (23%)36 (26%)13 (18%)0.2
Pacemaker implantation before 199483 (39%)29 (21%)54 (75%)<0.001
Indication for pacemaker
 Acquired sick sinus syndrome78 (37%)45 (32%)33 (46%)0.05
 Acquired atrioventricular block69 (33%)51 (37%)18 (25%)0.09
 Surgical atrioventricular block60 (28%)41 (29%)19 (26%)0.6
 Surgical sick sinus syndrome4 (2%)2 (1%)2 (3%)0.6
Pacing mode
 AAI(R)14 (7%)14 (10%)<0.001
 DDD(R)120 (57%)120 (86%)
 VDD(R)5 (2%)5 (4%)
 VVI(R)72 (34%)72 (100%)
Implantation technique
 Endocardial leads141 (67%)104 (75%)37 (51%)<0.001
 Epicardial leads70 (33%)35 (25%)35 (49%)
  • CHD, congenital heart disease.

  • aExpressed as median with IQR.

Figure 1

Congenital heart disease (CHD) diagnosis stratified by pacing mode at initial implantation. Ao valve, aortic valve disease; ASD, atrial septal defect; AVSD, atrioventricular septal defect; cc-TGA, congenital corrected transposition of the great arteries; Coarctation, coarctation of the aorta; Ebstein, Ebstein's anomaly; PAPVR/TAPVR, partial or total anomalous pulmonary venous return; Pul valv, pulmonary valve disease; SV, single ventricle physiology; TGA, Transposition of the Great Arteries; ToF, Tetralogy of Fallot; VSD, ventricular septal defect.

Atrial arrhythmias during follow-up

During a median follow-up of 13 years, 90 patients (43%) developed atrial arrhythmias. Of these 90 patients, 58 patients (64%) had intra-atrial reentrant tachycardia/atrial tachycardia and 32 patients (36%) had atrial fibrillation as the first presenting atrial arrhythmia. The cumulative rate of atrial arrhythmias was 21, 29, and 39% at 5, 10, and 15 years, respectively. Univariate predictors of atrial arrhythmias are depicted in Table 2. Atrial-based pacing, history of atrial arrhythmias, older age (≥18 years) at pacemaker implantation, sick sinus syndrome, complex CHD, and endocardial pacing system were identified as significant univariate predictors of atrial arrhythmias. Multivariate analysis, including all significant univariate predictors, demonstrated that only a history of atrial arrhythmias, older age (≥18 years) at pacemaker implantation, and complex CHD were independent predictors of atrial arrhythmias. Atrial-based pacing was not an independent predictor of sustained atrial arrhythmias during follow-up [hazard ratio (HR), 1.53; 95% confidence interval (CI), 0.91–2.56; P = 0.1] (Figure 2).

View this table:
Table 2

Predictors of atrial arrhythmia in CHD patients receiving pacemaker therapy

VariableUnivariate analysisMultivariate analysis
Hazard ratio (95% CI)P valueHazard ratio (95% CI)P value
History of atrial arrhythmia6.90 (4.40–10.8)<0.00015.55 (3.47–8.89)<0.0001
Older age (≥18 years) at implant2.74 (1.71–4.40)<0.00012.29 (1.29–4.04)0.005
Sick sinus syndrome1.99 (1.32–3.01)0.0011.50 (0.96–2.36)0.08
Atrial-based pacing1.89 (1.19–3.01)0.0071.53 (0.91–2.56)0.1
Complex CHD1.67 (1.10–2.54)0.021.57 (1.01–2.45)0.04
Endocardial pacemaker1.66 (1.06–2.62)0.021.10 (0.65–1.86)0.7
Female gender0.71 (0.47–1.08)0.1
  • CHD, congenital heart disease.

Figure 2

Freedom from atrial arrhythmias for ventricular pacing and atrial-based pacing in patients with CHD. Atrial, atrial-based pacing; HRadj, adjusted hazard ratio, HRunadj, unadjusted hazard ratio; ventricular, ventricular pacing.

Subgroup analysis

Subgroup analysis demonstrates that patients with single ventricle physiology had a high prevalence of previous atrial arrhythmias (50%) at pacemaker implantation (Table 3). Interestingly, no patient with aortic valve disease was known with a history of atrial arrhythmia prior to implantation. During follow-up, patients with a single ventricle physiology were more prone to develop atrial arrhythmias compared with the other groups (Figure 3, Table 3).

View this table:
Table 3

Overview of pacemaker details and atrial arrhythmias in selected CHD groups

VariableComplete TGAa (n = 35)Tetralogy of Fallot (n = 26)Aortic valve disease (n = 26)ASD secundum (n = 25)Single ventricle physiology (n = 18)P value
Clinical and pacemaker data
 Age at first implantation (years)b18 (8–25)28 (15–40)34 (18–51)27 (15–45)24 (13–29)
 Male gender69%62%73%36%44%0.03
 First implant <18 years49%39%23%32%33%0.33
 History of atrial arrhythmias29%23%0%28%50%0.004
 Atrial-based pacing54%54%77%60%88%0.05
 Epicardial leads46%12%15%28%72%0.07
Follow-up data
 5-year rate of atrial arrhythmias20%25%0%21%51%
 10-year rate of atrial arrhythmias32%30%11%27%60%
  • TGA, transposition of the great arteries; ASD, atrial septal defect.

  • aAtrial repair.

  • bExpressed as median with IQR.

Figure 3

Freedom from atrial arrhythmias in selected CHD groups after pacemaker implantation. Ao valve disease, aortic valve disease; ASD-II, secundum atrial septal defect; SV, single ventricle disease; ToF, Tetralogy of Fallot; TGA, Transposition of the Great Arteries.

Peri-procedural complications

Overall, the peri-procedural complication rate is comparable between patients with atrial-based pacing and those with ventricular pacing (Table 4). When looking at the different complications, there was a higher incidence of pocket infection in those with ventricular pacing when compared with those with atrial-based pacing (5.6 vs. 0%, P = 0.01).

View this table:
Table 4

Peri-procedural complication rates for atrial-based and ventricular pacing

ComplicationTotal (n = 211)Atrial-based (n = 139)Ventricular (n = 72)P value
Total number of patients22 (10.4%)11 (7.9%)11 (15.3%)0.1
 Lead failure9 (4.3%)5 (3.6%)4 (5.6%)0.5
 Bleeding3 (1.4%)3 (2.2%)0.5
 Pneumothorax2 (0.9%)2 (1.4%)0.5
 Pocket infection4 (1.9%)4 (5.6%)0.01
 Ventricular arrhythmias2 (0.9%)2 (2.8%)0.1
 Other2 (0.9%)1 (0.7%)1 (1.4%)1.0


In the general population without CHD, atrial-based pacing is beneficial in reducing the incidence of atrial fibrillation compared with ventricular pacing.510 The present multicentre study, however, demonstrates that the beneficial effects of atrial-based pacing cannot be extrapolated to the adult CHD population. After correction for important confounders, the incidence of atrial arrhythmias was not different between atrial-based pacing and ventricular pacing. These findings suggest that the mode of pacing does not play an important role in adults with CHD for the reduction of atrial arrhythmias.

Patients with CHD are prone to rhythm and conduction disease, which is not surprising considering the underlying congenital cardiac defect, abnormal anatomy, longstanding haemodynamic alterations, and the consequences of surgical interventions. Atrial arrhythmias are a burden to the adult CHD population and are associated with increased morbidity and mortality.24 These arrhythmias may be difficult to treat pharmacologically and may require (a combination of) catheter ablation or atrial antitachycardia pacing.16,17 The purpose of the present study was to explore whether the pacing mode may play a role in the prevention of atrial arrhythmias in CHD patients who had a bradycardia indication for permanent pacing.

Loss of AV synchrony due to ventricular pacing may contribute to increased atrial arrhythmogenesis. A meta-analysis by Healey et al.,5 summarizing the data of five randomized trials in patients without CHD, demonstrated that atrial-based pacing reduced the incidence of atrial fibrillation (HR, 0.80; 95% CI, 0.72–0.89; P< 0.001) and stroke (HR, 0.81; 95% CI, 0.67–0.99, P = 0.035) compared with ventricular pacing.

Interestingly, our study cohort showed no beneficial effect of atrial-based pacing in preventing atrial arrhythmias. Unexpectedly, we found that atrial-based pacing was associated with an increased risk of atrial arrhythmias in the univariate analysis (HR, 1.89; 95% CI, 1.19–3.00, P = 0.006). This can be explained by the older age at pacemaker implantation of the cohort who received atrial-based pacing. In the multivariate analysis, the pacing mode was not a predictor of atrial arrhythmias. One previous single-centre study demonstrated that atrial-based pacing did not appear to be protective against subsequent atrial arrhythmias.13 The reason for this lack of beneficial effect of atrial-based pacing may be inherent to the increased vulnerability of the adult CHD population for atrial arrhythmias due to increased natriuretic peptides, altered loading conditions, and presence of surgical atrial scar.18,19 The loss of AV synchrony induced by ventricular pacing may add little incremental arrhythmogenic risk in such patients, and therefore atrial-based pacing may not be as beneficial as in the general population. This is in agreement with results from the Mode Selection Trial (MOST) in Sinus-Node Dysfunction, where a significant reduction in atrial fibrillation with DDDR pacing when compared with ventricular pacing was observed solely in the subgroup of patients without a history of atrial fibrillation.9 These results support that pacing mode selection is less important with respect to later development of atrial fibrillation in patients who already are at high risk of developing atrial fibrillation.

Another explanation for the lack of beneficial effect of atrial-based pacing may be related to the type of atrial arrhythmias in the CHD population. In the present study, the majority of patients with atrial arrhythmias during follow-up experienced intra-atrial reentry tachycardia. Approximately one-third of patients had atrial fibrillation. This ratio is in accordance with a previous study in adult CHD patients who underwent electrical cardioversion.20 Atrial-based pacing may be less beneficial for the prevention of intra-atrial reentrant tachycardia than atrial fibrillation.

With regard to predictors for future arrhythmias after pacemaker implantation, Harris and coworkers13 identified a history of atrial arrhythmias as an independent predictor. In addition, we identified older age (≥18 years) at pacemaker implantation and complex CHD as independent predictors of atrial arrhythmias. In particular, patients with single ventricle physiology, mostly Fontan patients with an atriopulmonary connection, are at increased risk of developing atrial arrhythmias later in life (Figure 3). Previous electrophysiological mapping studies have shown that Fontan patients undergo progressive adverse atrial mechano-electrical remodelling with increasing age, rendering them more prone to atrial arrhythmias.19

It is important to stress that the primary aim of the current study was to investigate the role of pacing mode on the incidence of atrial arrhythmias. This study does not evaluate the potential beneficial effects of atrial-based pacing, by maintaining AV synchrony, on haemodynamic parameters, risk of heart failure, and avoidance of symptoms related to pacemaker syndrome.

The current study demonstrated a relative high proportion of patients who received ventricular pacing as the initial pacing mode (34%). Implantation of the pacemaker in an older era (before 1994) or at paediatric/adolescent age may explain this phenomenon. The beneficial effects of dual-chamber pacing became more appreciated from the early 1990s.7,8 However in the paediatric population, there is still some debate whether dual-chamber pacing is superior to ventricular pacing.21

Finally, the rate of peri-procedural complications in the general population without CHD is nearly twice as high with atrial-based pacing compared with ventricular pacing (6.2 vs. 3.2%), primarily due to a significant increase in the rate of lead failure and infection.5 Our study cohort did not confirm this difference in overall peri-procedural complication rate between patients with atrial-based pacing or ventricular pacing. However, CHD patients with ventricular pacing did demonstrate more pocket infections (5.6 vs. 0%, P = 0.01).

Study limitations

This study has limitations inherent to any retrospective study. The major limitation of the present study is the observational study design, which renders the data subject to selection bias. We tried to overcome this limitation by using a multivariate analysis including all potential confounders. Clearly, this approach cannot replace a randomized controlled trial design. Furthermore, despite its relative large size due to its multicentre nature, this study may be underpowered to demonstrate a potential benefit of atrial-based pacing. In addition, the heterogeneity of pacemaker settings limited our ability to draw conclusions with regard to desired pacemaker settings. Owing to the retrospective study design, we had incomplete echocardiographic data limiting our ability to correct for potential confounders such as atrial dilatation or significant atrioventricular regurgitation. Finally, the patient sample is drawn from an adult-based practice and hence skewed towards this age group. Therefore, the patients who died or were lost to follow-up before reaching adulthood are not addressed in this study. Considering the abovementioned limitations, the conclusions of the present study must be drawn with caution.


Compared with ventricular pacing, the use of atrial-based pacing was not associated with a lower incidence of atrial arrhythmias in adults with CHD. There is a high vulnerability for developing atrial arrhythmias, especially intra-atrial reentrant tachycardia, which is largely determined by older age, underlying atrial substrate, and prior atrial arrhythmias. Atrial-based pacing may be less beneficial for prevention of intra-atrial reentrant tachycardia than atrial fibrillation. To develop effective therapies for preventing atrial arrhythmias, future research should focus on understanding the pathophysiology of atrial arrhythmia in adults with CHD.

Conflict of interest: none declared.


View Abstract