OUP user menu

Effects of right low atrial septal vs. right atrial appendage pacing on atrial mechanical function and dyssynchrony in patients with sinus node dysfunction and paroxysmal atrial fibrillation

Mei Wang, Chung-Wah Siu, Kathy LF Lee, Wen-Sheng Yue, Guo-Hui Yan, Stephen Lee, Chu-Pak Lau, Hung-Fat Tse
DOI: http://dx.doi.org/10.1093/europace/eur110 1268-1274 First published online: 21 April 2011

Abstract

Aims To study the effects of right low atrial septum (AS) and right atrial appendage (RAA) pacing on atrial mechanical function and dyssynchrony in patients with sinus node disease (SND) and paroxysmal atrial fibrillation (AF).

Methods and results Detailed echocardiographic examination was performed on 30 patients with SND and paroxysmal AF and a dual-chamber, dual sensing, dual response pacemaker with atrial lead implantation at AS(n= 15) or RAA(n= 15). Peak atrial velocities were recorded by pulsed tissue Doppler spectrum. The timing of atrial contractions (Ta) was measured at the middle of the left atrial (LA) and right atrial (RA) free wall. Intra-[standard deviation (SD) of time of Ta (Ta-SD)] and inter-atrial delay(Ta-RL) was measured as the SD of time interval among LA six segments and time difference between the LA and RA wall, respectively.

The baseline clinical statuses were similar between groups. Indexes of LA function, and intra- or inter-atrial dyssynchrony were also similar during intrinsic sinus rhythm in both groups (all P> 0.05). During atrial pacing, LA ejection fraction (52 ± 16 vs. 39 ± 14%, P= 0.029) and LA active emptying fraction (34 ± 7 vs. 23 ± 15%, P= 0.012) were higher in patients with AS than RAA pacing. Atrial velocity was also higher at the RA free wall (14.3 ± 3.1 vs. 10.3 ± 4.4 cm/s, P= 0.009), LA septal (7.5 ± 2.1 vs. 5.2 ± 1.7 cm/s, P= 0.004) and lateral wall (8.6 ± 2.4 vs. 6.3 ± 3.0 cm/s, P= 0.024) during AS compared with RAA pacing. There was no difference in Ts-SD during atrial pacing, nevertheless Ta-RL was significantly prolonged in patients with RAA compared with those with AS pacing (42 ± 36 vs. 27 ± 25 ms, P= 0.011).

Conclusion In patients with SND and paroxysmal AF, right low AS pacing significantly improved global and regional atrial mechanical function and synchronized inter-atrial electromechanical contraction compared with RAA pacing.

  • Atrial septum pacing
  • Appendage pacing Atrial dyssynchrony
  • Sinus node disease
  • Paroxysmal atria fibrillation

Introduction

Sinus node dysfunction (SND) is a major indication for permanent cardiac pacemaker implantation,1 and is associated with an increased risk of atrial fibrillation (AF).2 Although atrial-based dual-chamber pacing in patients with SND reduces the incidence of AF,35 long-term follow-up reveals that 22–28% of patients with SND develop persistent AF after dual-chamber pacemaker implantation.2 The occurrence of AF episodes following pacemaker implantation is a powerful, independent predictor for major cardiovascular events, including stroke.6,7 Nevertheless, the pathophysiology of AF in patients with SND remains unclear. It has been postulated that changes in atrial electrical and mechanical properties related to atrial myopathy in SND contribute to the development of AF.8 Indeed, prior clinical studies have shown that atrial conduction delay is observed in up to one-third of patients with SND and AF.9 Our recent study 10 showed that the presence of reduced active atrial contraction and prolonged inter-atrial electromechanical dyssynchrony in patients with SND is independently correlated with the occurrence of paroxysmal AF.

Previous clinical studies demonstrated that atrial septal (AS) pacing is associated with shortened inter-atrial conduction delay and decreased P-wave duration,11 and reduces the recurrence of AF in patients with bradycardia compared with traditional right atrial appendage (RAA) pacing.1214 The effects though of AS pacing vs. RAA pacing on atrial electromechanical properties in patients with SND and paroxysmal AF remain unclear. The purpose of this study was to compare the atrial mechanical function and dyssynchrony in patients with SND and paroxysmal AF who underwent permanent right low AS pacing vs. RAA pacing.

Methods

Study population

Consecutive patients with symptomatic SND and paroxysmal AF who underwent dual-chamber pacemaker implantation between April 2002 and February 2008 at our pacemaker clinic were screened. All patients had a symptomatic sinus bradycardia or prolonged sinus pause >3 s. The presence of paroxysmal AF was defined as any AF episode documented by 12-lead electrocardiogram (ECG) or Holter recording prior to device implantation. All patients were implanted with an active atrial lead, located at the right low AS or RAA, and a passive ventricular lead located at the right ventricular apex (Figure 1). In patients who received right low AS pacing, the atrial active lead was positioned at the Triangle of Koch region as described previously 15,16 using a custom-shaped style. Patients with a documented history of congestive heart failure, valvular heart disease, and cardiomyopathy, persistent AF episodes within last 48 h, intrinsic sinus rate <45 bpm, or concomitant high degree atrioventricular block were excluded from study. As a result, a total of 15 patients with SND and right low AS pacing who consented for this study were initially enrolled, and then 15 age- and sex-matched patients with SND and RAA pacing were also recruited for comparison.

Figure 1

Fluoroscopic images at right anterior oblique (RAO, upper panel) and left anterior oblique (LAO, lower panel) in patients with right low atrial septal (AS) and right atrial appendage (RAA) pacing.

Study design

Baseline demographic characteristics, clinical status, medical therapy, and device memory diagnostic counters of all patients were retrieved from medical records and device, respectively. According to atrial lead position, patients were divided into two groups: group 1: right low AS pacing (n= 15); and group 2: RAA pacing (n= 15). All patients underwent detailed echocardiographic examination using tissue Doppler imaging during atrial pacing and sinus rhythm. During the atrial pacing mode, pacemakers were programmed at atrial pacing, atrial sensing, inhibition response mode with lower pacing rate set at 60 or 10 bpm above the intrinsic sinus rhythm to ensure continuous atrial pacing for 1h prior to echocardiographic examination for atrial and ventricular function. A second echocardiographic examination of atrial function was performed 1h later in patients who had an intrinsic sinus rhythm rate of at least 45 bpm after reprogramming their pacemakers to backup ventricular pacing, ventricular sensing, inhibition response mode at 40 bpm.

Echocardiographic examination was only performed during the follow-up visit when no AF episode was detected by pacemaker diagnostic counter for at least 48 h prior to pacemaker reprogramming. The AF burden in the last 3 months was calculated from device memory diagnostic counters for atrial high-rate episodes, when atrial rate exceeded 180 bpm for 6 min.16 The maximum P-wave durations as recorded from 12-lead surface ECG were measured during intrinsic sinus rhythm and atrial pacing in all subjects. This study was approved by the local Institutional Review Board and informed consent was obtained from all participants.

Echocardiography and tissue doppler imaging

Detailed echocardiographic examination with tissue Doppler imaging was performed in all patients to determine atrial mechanical function and dyssynchrony. An IE33 echocardiographic machine (Philips Medical Systems, Bothell, WA USA) with an S5-1 transducer was used with the patient in the left decubitus position and a single experienced cardiologist (G.H.Y.) blinded to all clinical details of patients. Standard two-dimensional and M-mode measurements were obtained according to the guidelines of the American Society of Echocardiography,17 and left ventricular ejection fraction (LVEF) was measured using modified biplane Simpson's method from apical four- and two-chamber views. Mitral inflow E velocity and its deceleration time, and mitral inflow A velocity were measured using pulse wave Doppler with the sample volume placed at the tips of the mitral valve from the apical four-chamber view.

Pulsed tissue Doppler imaging spectrum was obtained for long-axis motion from the apical four-chamber view as described previously.10 Angle, depth, and pulse repetitive frequency were adjusted to accommodate a possible highest frame rate (usually frame rate >150 frames/s). Colour-coded tissue Doppler imaging loop for zoomed left atrium (LA) and right atrium (RA) were also stored. At least three consecutive beats were analysed off-line using a workstation (Q Lab, Version 4.0, Philips Ultrasound).

The ratio of mitral inflow E velocity to myocardial Esep velocity (E/Esep) measured by pulsed tissue Doppler spectrum was calculated as an index of LV filling pressure. Diastolic function was categorized as: normal filling pattern (mitral inflow E/A ratio>0.5, deceleration time >220 ms < 280 ms, E/Esep < 8), abnormal relaxation filling pattern (mitral inflow E/A ratio <0.5, deceleration time >280 ms, E/Esep > 12), pseudonormal or restrictive filling pattern (mitral inflow E/A ratio >1.5, deceleration time <220 ms, E/Esep > 15).18

Several parameters were used to assess LA haemodynamic performance. Left ventricular diameter was measured at the parasternal long-axis view at the end of systole. At apical four-chamber view, LA volume (LAV) at LV end-systole (ESV), LAV at the ECG onset of P wave (PV), and LAV at end-diastole (EDV) were traced based on the ECG signal. LA volume index was calculated using ESV/body surface area.10 Left ventricular ejection fraction was calculated using modified Simpson's method by measuring LAV at the end of systole minus the end of diastole divided by LAV at the end of systole: (ESV – EDV) / ESV × 100%. Left ventricular active emptying fraction represents the contribution of atrial systole to the filling of LV, measured by LAV at the onset of P wave minus end of diastole, then divided by LAV at onset P wave: (PV – EDV) / PV × 100%. Left ventricular filling fraction was calculated by A waveVTI/mitralVTI, respectively.19,20

Atrial contraction velocities were measured by pulsed tissue Doppler spectrum on the annulus of the RA free wall (Arv), inter-AS wall (Asep) and LA lateral wall (Alat). The time interval (Ta) profile of atrial myocardial wall contraction was obtained by tissue Doppler imaging, placing the sample volume at the middle of the RA free wall and LA six segments (septal, lateral, inferior, anterior, posterior, and anterior-septal wall) from apical four-chamber, two-chamber, and apical long-axis views. Ta was measured as the time interval from peak R wave of QRS complex to peak atria contraction velocity. Intra-atrial synchronicity was assessed as the standard deviation (SD) of time of Ta (Ta-SD) among six segments of LA or time difference between LA septal and lateral wall (Ta-SL). The degree of inter-atrial synchronicity (Ta-RL) was determined by the time difference between Ta at the RA free wall and LA lateral wall (Figure 2).

Figure 2

Echocardiographic images with tissue Doppler imaging in patients with right low atrial septal (AS) and right atrial appendage (RAA) pacing. Examples of tissue Doppler imaging in patients in Sinus node dysfunction and paroxysmal atrial fibrillation during right low atrial septal pacing (A) and RAA pacing (B). Up left graph is zoomed left and right atrium with tissue Doppler imaging on apical four-chamber view. Up right graph is tissue Doppler colour cording imaging showing atrial contraction synchronized (white arrow). Low panel graph is the curves of left and right atrial free wall contraction by tissue Doppler spectrum. The red line represents right atrial free wall, the blue line represents left atrial free wall spectrum.

In this study, all the off-line data analysis was performed by a single experienced cardiologist (M.W.) who was blinded to the clinical status of study subjects using a workstation. Specially, the intra-observer variability of atrial tissue Doppler measurements was determined in five randomly selected subjects from each group.

Statistical analysis

When calculating the sample size, we assumed the SD of the LAEF to be 11% 10 and the same among patients and controls. Based on the hypothesis that LAEF fraction is higher among patients with right low AS pacing, we used a two-sided test with a 0.05 level of significance. In order to detect a difference of 15% in the means of the two groups with 95% power, a sample size of n = 15 in each group was required.

Data are presented as percent or mean ± SD as appropriate. Comparisons were made between categorical variables using χ2 test. Comparisons between patients with right low AS pacing and RAA pacing during intrinsic sinus rhythm and atrial pacing were made using an independent t test. Comparisons between changes in intrinsic sinus rhythm and atrial pacing in patients with AS and RAA were made using paired t test. All statistical analyses were performed with SPSS software package (version 16.0, SPSS Inc., Chicago, IL, USA). A P value of <0.05 was considered statistically significant.

Results

Clinical characteristics and echocardiographic examination of left ventricular function

The baseline demographic features of the study population are shown in Table 1. There were no significant differences in age, gender, prevalence of hypertension, diabetes, coronary artery disease, medications, or duration of pacing between the right low AS pacing group, and RAA pacing group (all P> 0.05). Their respective device diagnostic counters also showed similar AF burden and percentage time of atrial and ventricular pacing during the prior 3 months before the echocardiographic examination (Table 1, all P> 0.05). As measured from surface ECG, maximum P-wave duration of ECG was significant shorter in patients with low AS pacing (79 ± 10 ms) than their baseline without pacing (143 ± 23 ms, P< 0.001), and those in patients with RAA pacing (106 ± 22, P= 0.001; Table 1).

View this table:
Table 1

Clinical characteristics of patients with right low atrial septal pacing and right atrial appendage pacing

Low AS (n= 15)RAA (n= 15)P value
Age, years70 ± 875 ± 90.16
Female gender12121.00
Duration of pacing, months23 ± 1724 ± 160.77
Hypertension, n (%)12 (80)11 (73)0.50
Diabetes mellitus, n (%)6 (40)6 (40)0.65
Hyperlipidaemia, n (%)4 (27)2 (13)0.33
Coronary heart disease, n (%)5 (33)5 (33)0.65
Maximum P-wave duration, ms
Sinus rhythm143 ± 23138 ± 170.67
Atrial pacing79 ± 10106 ± 220.001
Heart rate, bpm64 ± 864 ± 70.65
Atrial fibrillation burden8.6 ± 2215.2 ± 330.53
Medications, n (%)
Amiodarone or sotalol6 (40)6 (40)0.65
Aspirin or warfarin13 (87)13 (87)0.70
ACEI or ARB5 (33)2 (13)0.20
Calcium channel blockers7 (47)4 (27)0.23
Beta-blockers8 (53)6 (40)0.36
Statin4 (27)3 (20)0.50
Device diagnostic counter, %
Percentage of atrial pacing86 ± 3168 ± 340.25
Percentage of ventricular pacing14 ± 3232 ± 380.27
Echocardiographic examination of LV systolic and diastolic function
LV ejection fraction, %62 ± 556 ± 120.12
E/Esep15 ± 515 ± 60.94
Normal diastolic pattern, %9 (60)7(47)0.36
Abnormal or pseudonormal diastolic pattern, %6 (40)8 (53)0.50
Pulmonary artery systolic pressure, mmHg31 ± 738 ± 120.10
  • ACEI, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker; E/Esep, ratio of mitral inflow E velocity to septal mitral annulus myocardial velocity by tissue Doppler spectrum.

Echocardiographic assessment of LV systolic and diastolic function during atrial pacing showed no significant differences in LVEF or LV mass or the prevalence of diastolic dysfunction or E/Esep between the two groups (Table 1, all P> 0.05).

Left atrial mechanical function and dyssynchrony

Echocardiographic parameters of LA mechanical function and dyssynchrony during intrinsic sinus rhythm and atrial pacing are shown in Table 2.

View this table:
Table 2

Echocardiographic parameters of left atrial mechanical function and dyssynchrony in patients with atrial septum and right atrial appendage pacing during intrinsic sinus rhythm and atrial pacing

Intrinsic sinus rhythmAtrial pacing rhythm
Low AS (n= 15)RAA (n= 15)P valueLow AS (n= 15)RAA (n= 15)P value
Left atrial diameter, cm3.8 ± 0.73.9 ± 0.80.553.9 ± 0.64.0 ± 0.90.57
Left atrial volume index, ml/m239 ± 1739 ± 150.9942 ± 1547 ± 180.64
Left atrial ejection fraction, %54 ± 950 ± 170.5652 ± 1639 ± 140.029
Left atrial active emptying fraction, %24 ± 1224 ± 180.6734 ± 7*23 ± 150.012
Left atrial filling fraction, %32 ± 731 ± 110.8136 ± 6*39 ± 11*0.72
Arv, cm/s13.2 ± 5.311.0 ± 3.10.3314.3 ± 3.110.3 ± 4.40.009
Asep, cm/s6.2 ± 1.95.2 ± 1.80.327.5 ± 2.15.2 ± 1.70.004
Alat, cm/s6.8 ± 2.36.2 ± 2.70.678.6 ± 2.46.3 ± 3.00.024
Ta-SD, ms32 ± 2556 ± 560.2923 ± 1120 ± 150.60
Ta-SL, ms24 ± 1828 ± 370.8628 ± 1837 ± 360.38
Ta-RL, ms38 ± 2624 ± 170.2527 ± 2542 ± 360.011
  • *P < 0.05 when compared with intrinsic sinus rhythm pacing of each mode.

  • Arv, atrial contraction velocity on annulus of right ventricular free wall; Asep, atrial contraction velocity on annulus of left ventricular septal wall; Alat, atrial contraction velocity on annulus of left ventricular lateral wall; Ta-SD, the standard deviation of intra-atrial dyssynchrony by tissue Doppler imaging; Ta-SL, the time delay of left atrial septal to lateral wall; Ta-RL, the time difference between Ta at right atrial free wall and left atrial lateral wall.

During sinus rhythm, there were no differences in LA dimension, LAV index, LAEF or LA active emptying fraction, and atrial contraction velocities among Arv, Asep, and Alat between the two groups (P> 0.05). The degree of intra- and inter-atrial dyssynchrony as expressed by Ta-SD, Ta-SL, and Ta-RL were also similar between the two groups during sinus rhythm (all P> 0.05).

During atrial pacing rhythm, there were no significant differences in LA dimension or LAV index between the two groups (Table 2, P> 0.05). Nonetheless, the LAEF (52 ± 16 vs. 39 ± 14%, P= 0.029, Figure 3A) and LA active emptying fraction (34 ± 7 vs. 23 ± 15%, P= 0.012, Figure 3B) were significantly higher in patients with right low AS pacing compared with those with RAA pacing.

Figure 3

Changes in (A) left atrial (LA) ejection fraction; (B) left atrial active emptying fraction; (C) left atrial filling fraction; (D) atrial contraction velocity on annulus of right ventricular free wall (Arv); (E) atrial contraction velocity on annulus of left atrial septal wall (Asep); (F) atrial contraction velocity on annulus of left atrial lateral wall (Alat); (G) intral-atrial dyssynchrony (Ta-SD); and (H) inter-atrial dyssynchrony (Ta-RL) from intrinsic sinus rhythm to atrial pacing in patients with right low atrial septal (AS) and right atrial appendage (RAA).

The intra-observer variability for Ta were 7.8% for Ta from R wave to RA free wall, 5.8% for Ta from R wave to septal wall, and 10% for Ta from R wave to LA lateral wall, respectively. As determined by tissue Doppler imaging, the atrial myocardial velocities among Arv (14.3 ± 3.1 vs. 10.3 ± 4.4 cm/s, P= 0.009, Figure 3D), Asep (7.5 ± 2.1 vs. 5.2 ± 1.7 cm/s, P= 0.004, Figure 3E), and Alat (8.6 ± 2.4 vs. 6.3 ± 3.0 cm/s, P= 0.024, Figure 3F) were also significantly higher in patients with right low AS pacing compared with RAA pacing group during atrial pacing. There was no significant difference in Ts-SD between the two groups during atrial pacing (Figure 3G); however, Ta-RL was significantly prolonged in patients with RAA pacing (42 ± 36 vs. 27 ± 25 ms, P= 0.011, Figure 3H).

Comparison between sinus rhythm and atrial pacing rhythm showed no significant differences in LAEF, atrial myocardial velocities, and the degree of intra- and inter-atrial dyssynchrony among patients with right low AS pacing or RAA pacing (Table 2, all P> 0.05). Nevertheless, the LA active emptying fraction during right low AS pacing, and the LA filling fraction (Figure 3C) during both right low AS pacing and RAA pacing were also significantly augmented compared with those during intrinsic sinus rhythm (Table 2, P< 0.05).

Discussion

In this study, detailed echocardiographic examination was performed to assess LA haemodynamic function, mechanical contraction, and dyssynchrony in patients with SND and paroxysmal AF who received either right low AS pacing or RAA pacing. There were no significant differences in clinical characteristics, device diagnostic parameters or LV systolic and diastolic function between the two groups. During intrinsic sinus rhythm, the indexes of LA haemodynamic function and mechanical contraction, as well as intra- or inter-atrial dyssynchrony were similar between patients with right low AS and RAA pacing. Nonetheless, during atrial pacing, right low AS pacing was associated with improved LA haemodynamic function such as LAEF and LA active emptying fraction, and increased active atrial myocardial contraction velocity on annulus of right free wall, septal wall, and lateral wall compared with RAA pacing. Furthermore, right low AS pacing significantly reduced Ta-RL, suggesting improved inter-atrial synchronization with right low AS pacing in patients with SND.

Previous studies have shown that several factors, such as bradycardia, delayed inter- and/or intra-atrial conduction, atrial tissue mass, atrial stretch, and the interaction with the autonomic nervous system may affect the occurrence of AF.21 In patients with SND, conventional atrial pacing at the RAA may induce delayed activation of areas with atrial substrate that is important in the initiation of paroxysmal AF.9 Several electrophysiological studies have shown that in patients with paroxysmal AF, single-site pacing at either the inter-AS or Bachmann's bundle is more effective than biatrial or dual-site atrial pacing in the prevention of AF induction by shortening atrial activation times,1315,2224 and avoiding the undesirable prolongation of the inter-atrial conduction in patients with AF.2528 In this study, we confirmed that in patients with right low AS pacing there was significantly shorter P-wave duration compared with RAA pacing.

Our previous results 10 demonstrated that patients with SND patients and paroxysmal AF have more reduced LA regional and overall mechanical contraction compared with those without AF, despite similar LAV, and LV systolic and diastolic function. The presence of a lower LA contraction velocity is associated with a reduced LA active emptying fraction. Furthermore, reduced active atrial contraction and prolonged inter-atrial dyssynchrony independently correlate with the presence of paroxysmal AF in patients with SND. The effects of different atrial pacing sites on atrial electromechanical function in patients with SND and paroxysmal AF nonetheless remain unclear. In an acute haemodynamic study,29 we showed that AS pacing reduced atrial mechanical work and improved atrial mechanical efficacy in patients without AF. In this study, we further demonstrated that right low AS pacing improved inter-atrial mechanical dyssynchrony in SND patients with paroxysmal AF. Moreover, regional atrial longitudinal myocardial contraction as measured by atrial myocardial velocity using tissue Doppler imaging at different regions, including right free wall, septal, and lateral wall also provide further insight into the effects of different atrial pacing sites on atrial mechanical function. As compared with RAA pacing, right low RA pacing was associated with better regional active atrial contraction velocities over the right and left atria, and overall haemodynamic LA performance. Although there were no differences in regional atrial contraction velocities between sinus rhythm and during atrial pacing in right low AS pacing, indices of global LA mechanical function including LA active emptying fraction and filling fraction were also significantly improved during right low AS pacing. These findings suggest that maintaining inter-atrial electromechanical synchronization with right low AS pacing in patients with SND may contribute to improved atrial mechanical function. However, whether this improvement and dyssynchrony with right low AS pacing in patients with SND and paroxysmal AF can prevent AF requires further study by ongoing clinical trial.16

Study limitation

First, the small sample size of the study population was a limitation as only a small proportion of patients received low right AS pacing at the discretion of the operators. Secondly, this was a cross-sectional cohort study of patients with an implanted pacemaker, and echocardiographic examination was not performed prior to device implantation. Thus, the potential long-term impact of atrial pacing sites on atrial mechanical function and progression of AF needed to be addressed by future randomized controlled trial.

Conclusion

Our data demonstrate that right low AS pacing in SND patients with paroxysmal AF who have a dual-chamber, dual sensing, dual response pacemaker achieve better regional right and left atrial active mechanical properties and LA haemodynamic performance compared with those with RAA pacing. Inter-atrial electromechanical dyssynchrony was also reduced with right low AS pacing. Whether these improvements in atrial mechanical function and dyssynchrony associated with long-term AS pacing can prevent the recurrence of AF in patients with SND remains unclear.

Funding

This study was supported by General Research Fund of Research Grant Council of Hong Kong (HKU 7777/07M and HKU 7775/08M).

Conflict of interest: none declared.

Acknowledgement

This study was supported by General Research Fund of Research Grant Council of Hong Kong (HKU 7777/07M and HKU 7775/08M).

References

View Abstract