Cardiac Rhythm Management
Articles Articles 2013 November

p-Wave Duration and Dispersion and Ascertaining the Optimal Site for Atrial Pacing

DOI: 10.19102/icrm.2013.041105


Mount Sinai Heart Institute at Mount Sinai Medical Center, Miami Beach, FL

PDF Download PDF

ABSTRACT.  The purpose of this study was to determine the optimal pacing site in different anatomical locations in the right atrium by evaluating the p-wave maximum duration (PMD) and p-wave dispersion (PWD). Two patient populations were recruited for this study: those undergoing device implantation, and those referred for electrophysiologic study. Pacing was done at the right atrial appendage (RAA), right atrial free wall, and right atrial septum. The PMD and PWD were determined in these three different walls. There were 33 patients enrolled in the study, four of which were excluded from the analysis. Thus, a total of 29 patients were included in the final analysis. There were no significant differences in PMD and PWD between groups at baseline. There was a statistically significant difference in PMD between sinus rhythm and pacing in both the RAA and the right free wall, p<0.001. The mean PWD during sinus rhythm, RAA pacing, right free wall pacing, and right atrial septal pacing was 20±13, 22±11, 27±23, and 15±9 ms, respectively. There was no significant difference in PWD between sinus rhythm and all three right atrial pacing sites; however, there was a statistically significant difference in PWD between right free wall and right atrial septum pacing, p<0.01. This analysis demonstrates that septal pacing offers the most favorable PMD and PWD during atrial pacing.

KEYWORDS. p-wave dispersion, p-wave maximum duration, right atrium, pacing.

The authors report no conflicts of interest for the published content.
Manuscript received September 11, 2013, Final version accepted October 15, 2013.

Address correspondence to: Orlando Santana MD, Director, Echocardiography Laboratory Columbia University Division of Cardiology, Mount Sinai Heart Institute, 4300 Alton Road Miami Beach, Florida 33140. E-mail:


Atrial fibrillation commonly occurs in patients with sinus node dysfunction. Such patients, when paced for bradycardia support, continue to have a relatively higher incidence of recurrent atrial fibrillation. Several factors appear to be associated with the onset of this arrhythmia in pacemaker patients. It is recognized that excess right ventricular pacing causes atrial fibrillation.1 Dual-chamber pacing is superior to the single ventricular pacing mode in preventing atrial fibrillation, but it is not free of such risk.2,3 Thus, the atrial paced rhythm, once timing-based ventricular pacing has been eliminated, continues to demonstrate an important incidence of atrial fibrillation. Recently, we reported that, with an increasing proportion of atrial paced beats, there was an increasing risk of incident atrial fibrillation.4

The electrocardiographic measurements of atrial activation time as determined by the p-wave maximum duration (PMD), and the inhomogeneity of atrial conduction as determined by the PWD, offer insight into the arrhythmogenicity of atrial pacing.5 There is evidence that these parameters are dependent upon the atrial pacing site.6 Thus, the site of atrial pacing vis-à-vis these parameters may affect the relative incidence of atrial fibrillation. The purpose of this study is to explore the potential arrhythmogenicity of the paced atrial beat in different anatomical locations in the right atrium, and secondarily to identify criteria that might guide atrial lead positioning towards more favorable intra-atrial locations.

Methods and patient selection

This was a prospective study that was approved by the Mount Sinai Medical Center Institutional Review Board, and took place at our institution between November 2010 and April 2012. All patients signed written consent. Two populations were recruited for this study: those undergoing device implantation with an atrial lead, and those referred for electrophysiology study with or without radiofrequency ablation. The subjects were included if undergoing primary device implantation or at upgrade of a single-chamber ventricular to a dual-chamber device. Both pacemaker and implantable cardioverter-defibrillator patients were included. All were required to be in sinus rhythm at the time of implantation, and were at least 20 years of age. Subjects were excluded if electrocardiographic data quality or lead stability during pacing precluded adequate data analysis.

Pacing protocol

Bipolar pacing was performed at three separate anatomic sites in the right atrium, either during atrial lead implantation or during electrophysiologic study. Pacing was performed at the right atrial appendage (RAA), right atrial free wall, and right atrial septum. The RAA was defined as the most apical portion of the right atrial appendage. The anatomical pacing site was confirmed fluoroscopically in the right anterior oblique, anterior–posterior and left anterior oblique views (Figure 1).


Figure 1: Fluoroscopic views of the three different anatomical pacing sites. Left side right anterior oblique at 46 degrees, right side left anterior oblique at 44 degrees. 1: right atrial appendage; 2: right atrial free wall; 3: right atrial septal wall. Arrow: atrial catheter during an electrophysiologic study.

Electrophysiologic study

Pacing was performed with either a 5F Josephson quadripolar catheter (St. Jude Medical, Minnesota, MN) with 5 mm electrode spacing or a 6F Polaris-X decapolar catheter (Boston Scientific, San Jose, CA) with 5 mm electrode spacing. The catheters were inserted via femoral access. The pacing cycle length was 600 ms with a pacing output of 2–5 mA at 2 ms pulse width via a commercially available electrophysiology stimulator, EP-work-mate system (St. Jude Medical). The electrophysiologic study was done prior to ablation and before the use of any medications.

Device implantation

All atrial leads were 7F bipolar active fixation leads inserted via cephalic or subclavian vein access (Medtronic, Minneapolis, MN). In each case, the lead was manipulated to each pacing site prior to final lead positioning. Pacing was performed at each site without the helix deployed. The pacing cycle length was 600 ms with a pacing output of 2–5 mV at 0.5 ms pulse width via a pacing system analyzer (Medtronic).

Electrocardiographic analysis

The PMD was determined by identifying the p-wave of greatest duration between all electrocardiogram (ECG) leads as measured from the onset to the termination of the p-wave. The PWD was defined as the difference between the p-waves of greatest and least duration on all ECG leads. The measurements were made during sinus rhythm and at all pacing sites after confirming at least five consecutive captured beats at the programmed pacing cycle length. For precise identification and recognition of the p-waves, the signals were acquired and analyzed with a commercially available electrophysiology system, EP-work-mate system (St. Jude Medical). During sinus rhythm, the p-wave duration was determined by the interval from the onset to offset of the p-wave. During atrial pacing, the p-wave duration was determined by the interval from stimulus artifact, to the offset of the p-wave signal (Figure 2).


Figure 2: p-Wave digital measurement. Calipers are placed from the beginning of the p-wave (sinus rhythm) or stimulation artifact (pacing) to the end of the p-wave. Notice the increase in duration in p-wave when pacing in the appendage compared with the right atrial septal wall. BL: base line p-wave in sinus rhythm, AP: appendage pacing; SEP: septal pacing.

Data analysis was conducted with a sweep speed of 150 mm/s, low pass filter of 40 Hz, high pass filter 0.5 Hz, and a gain of 0.2 mV/cm with the clipping feature off. All measurements were done by two investigators, who were blinded to the pacing sites, and had an interobserver variability correlation of 0.9. All analyses were performed on all 12 standard surface leads.

Statistical analysis

All continuous variables are reported as mean values±1 standard deviation, or median and interquartile range (IQR 25–75%) as appropriate. Parametric continuous variables were analyzed with independent t-test or repeated measures ANOVA with Bonferroni’s post hoc comparisons while non-parametric variables were analyzed with the Mann–Whitney or Friedman test with post hoc Dunn’s comparisons; all dichotomous variables were analyzed with Fisher’s Exact test, p-values <0.05 were considered statistically significant. Spearman coefficients were used to evaluate correlation between the measurements by each investigator, a correlation equal to or greater than 0.7 was considered acceptable. The analysis was conducted using GraphPad Prism 4 (GraphPad Software Inc., San Diego, CA).


A total of 33 patients enrolled in the study. Four patients were excluded from analysis due to poor digital signal quality. Thus, 29 patients were included in the final analysis. Of the 29 patients, six (21%) were taking antiarrhythmic medications (amiodarone or sotalol). There were 16 (55%) patients referred for electrophysiologic study, five (38%) for dual-chamber pacemaker implantation, four (31%) for dual-chamber defibrillator implantation, and four (31%) for cardiac resynchronization therapy defibrillator.

There were baseline differences noted between the two groups, with the patients in the electrophysiology study group having a significantly greater history of atrial flutter but not atrial fibrillation (Table 1). The device group displayed a greater incidence of amiodarone use and heart failure, and had a wider QRS on baseline ECG. Despite these differences, there were no significant differences in PMD and PWD between groups at baseline. Therefore, the data were pooled for further analysis.

Table 1: Patient baseline characteristics


The mean PMD during sinus rhythm, RAA pacing, right free wall pacing, and right atrial septal wall pacing was 102±19, 127±28, 133±29, and 110±23 ms, respectively. There was a statistically significant difference in PMD between sinus rhythm and pacing in both the right atrial appendage (95% CI 34.8–14.9; p<0.001) and the right free wall (95% CI 41.8–22; p<0.001). There was no significant difference in PMD between sinus rhythm and right septal wall pacing or between RAA and right free wall pacing (CI 18.6–1.23; p>0.05) (Figure 3). The mean PWD during sinus rhythm, RAA pacing, right free wall pacing, and right atrial septal pacing was 20±13, 22±11, 27±23, and 15±9 ms, respectively. There was no significant difference in PWD between sinus rhythm and all three right atrial pacing sites; however, there was a statistically significant difference in PWD between right free wall and right atrial septum pacing (CI 2.35–23.4; p <0.01) (Figure 4).


Figure 3 : Mean p-wave duration in milliseconds. Note the similarity of right atrial septal pacing to baseline. ms: milliseconds; BL: base line p-wave in sinus rhythm; A: appendage pacing; FW: free wall pacing; S: septal wall pacing.


Figure 4: Median, 25th to 75th percentile, and minimum-maximum for p-wave dispersion in milliseconds. Notice that septal pacing has lowest dispersion. ms: milliseconds; BL: base line p-wave in sinus rhythm; A: appendage pacing; FW: free wall pacing; S: septum pacing.


This study demonstrates that PMD with septal pacing is similar to that seen during sinus rhythm. Pacing from the other tested right atrial areas produced a greater PMD than both sinus rhythm and septal pacing. On the other hand, while PWD did vary depending on the site of atrial pacing, this did not significantly differ from PWD in sinus rhythm, perhaps due to the small sample size. PWD was, however, significantly lower with septal rather than free wall pacing. These data demonstrate that septal pacing minimizes both PMD and PWD, and leads to a testable hypothesis that this site may be optimal for the prevention of atrial fibrillation in patients that require atrial pacing.

Measures of atrial conduction time and inhomogeneity have been shown to predict atrial fibrillation in patients without pacemakers.7 Early investigations relied on signal-averaged measures of p-wave duration, electrical dispersion, and the presence of late potentials.8 Owing to the complexity of signal-averaged ECG, simpler methods utilizing 12-lead ECG have been developed.9 As stated earlier, several factors seem to be associated with a higher onset of atrial fibrillation in individuals with pacemakers. It is has been recognized that ventricular pacing is not benign; this modality can not only incite detrimental effects in cardiac function and mechanics, such as a decline in left ventricular function, development of mitral regurgitation, ventricular remodeling, and ventricular dyssynchrony, but pacing can also affect the electrical properties of myocardium, imparting an increased susceptibility to atrial arrhythmias.10,11 It is suggested that right ventricular pacing can promote atrial fibrillation by several mechanism.12,13 This risk is present not only in single ventricular pacing but also dual-chamber devices,416 and recent studies have demonstrated that indeed ventricular pacing increases the atrial fibrillation burden and can also promote atrial fibrillation progression.17

Now there is evidence that atrial pacing alone is associated with an increased incidence of atrial fibrillation, with risk increasing with the frequency of atrial pacing.4 This risk may be different depending on the anatomical location used for right atrial pacing. It was noted that the dispersion of atrial refractoriness and intra-atrial conduction delays can be more influenced when stimulation is performed at the high right atrium versus the coronary sinus.18 Atrial pacing can cause intra-atrial and interatrial conduction delay that can be manifest on the electrocardiogram with changes in morphology, increase in the signal-averaged p-wave duration and p-wave dispersion, and also the PMD and PWD.5,19,20 Interestingly, some studies have demonstrated that when pacing the interatrial septum at the level of the Bachmann’s bundle not only produce shortening of the p-wave, but also shortening of intra-atrial conduction delay.21,22 There are several methods of calculating p-wave dispersion.19 However, the most clinically applicable method has been defined as the difference between the maximum and the minimum p-wave duration on 12-lead ECG.9 This measurement has been validated as a predictor of atrial fibrillation.6 It has been recognized that higher values of PMD and PWD are predictive of an increased risk of atrial fibrillation in patients with dual-chamber pacemakers, and a PMD of 120 ms and PWD of 40 ms correspond to a sixfold and 12-fold increased risk for the occurrence of persistent atrial fibrillation, respectively.6 It is has also been noted that depending on the site of right atrial pacing, both the PMD and PWD can vary significantly, suggesting the risk of atrial fibrillation may be modifiable depending on the chosen lead placement.23 Some studies have shown that septal pacing gives the shortest p-wave duration, atrial activation time, and PWD compared with more conventional pacing sites, such as the right atrial appendage.2428 It has been noted lately that interatrial septum pacing also seems to be better than pacing at the appendage in preventing progression of atrial fibrillation.29

Another finding of the present study is that these two indices of arrhythmogenicity are similar in all groups of patients, including those with a previous history of atrial fibrillation/flutter and patients currently receiving antiarrhythmic drug treatment. This suggests that these parameters are robust predictors of atrial fibrillation regardless of atrial substrate, and may support the generalizability of our findings.


This is a small, single-center, study without clinical or electrocardiographic follow-up after the indexed procedure; therefore, association of lead position and risk of developing atrial fibrillation cannot be determined. Additionally, some of the patients were on antiarrhythmic medication that could have affected intra-atrial and interatrial conduction properties, and thus the PMD and PWD; however, this is unlikely since no significant difference was noted at baseline. In subjects undergoing pacing lead implantation, the pacing protocol was performed prior to helix deployment. This may require a higher pacing output than would be required clinically, which may create a larger virtual electrode and shorten the p-wave. Finally p-wave onset during atrial pacing was defined at the electrical pacing artifact, which could have introduced minor errors in calculating p-wave duration because of potential latency. However, by pacing at 600 ms this latency was most likely minimal.


This analysis demonstrates that septal pacing offers the most favorable profile of PMD and PWD with less dispersion, which may impact arrhythmogenicity. Additionally, baseline PMD and PWD do not appear to be affected by baseline atrial arrhythmias, antiarrhythmic use, or heart failure. Perhaps the right atrial septal wall should be considered the preferential anatomical location for chronic atrial pacing, if good lead stability and acceptable pacing thresholds can be achieved. Further investigation should focus on randomized trials with long-term follow-up of lead position in several different anatomical regions of the right atrium to determine if there truly is an effect on the incidence of atrial fibrillation in this population.


  1. Sweeney MO, Bank AJ, Nsah E, et al. Minimizing ventricular pacing to reduce atrial. fibrillation in sinus-node disease. N Engl J Med 2007; 357:1000–1008. [CrossRef] [PubMed]
  2. Amasyali B, Köse S, Aytemir K, et al. The effect of VVI pacing on P-wave dispersion in patients with dual-chamber pacemakers. Heart Vessels 2006; 21:8–12. [CrossRef] [PubMed]
  3. Feuer JM, Shandling AH, Messenger JC. Influence of cardiac pacing mode on the long-term development of atrial fibrillation. Am J Cardiol 1989; 64:1376–1379. [CrossRef] [PubMed]
  4. Elkayam LU, Koehler JL, Sheldon TJ, Glotzer TV, Rosenthal LS, Lamas GA. The influence of atrial and ventricular pacing on the incidence of atrial fibrillation: a meta-analysis. Pacing Clin Electrophysiol 2011; 34:1593–1599. [CrossRef] [PubMed]
  5. Ausubel K, Klementowicz P, Furman S. Interatrial conduction during cardiac pacing. Pacing Clin Electrophysiol 1986; 9:1026–1031. [CrossRef] [PubMed]
  6. Demir AD, Soylu M, Ozdemir O, et al. Determinants of persistent atrial fibrillation in patients with DDD pacemaker implantation. Pacing Clin Electrophysiol 2003; 26:719–724. [CrossRef] [PubMed]
  7. Dilaveris PE, Gialafos EJ, Sideris SK, et al. Simple electrocardiographic markers for the prediction of paroxysmal idiopathic atrial fibrillation. Am Heart J 1998; 135:733–738. [CrossRef] [PubMed]
  8. Kutarski A, Głowniak A, Szcześniak D, Ruciński P. Effects of different atrial pacing modes evaluated by intra-cardiac signal-averaged ECG. Cardiol J 2008; 15:129–142. [CrossRef] [PubMed]
  9. Dilaveris PE, Gialafos JE. P-wave dispersion: a novel predictor of paroxysmal atrial fibrillation. Ann Noninvasive Electrocardiol 2001; 6:159–165. [CrossRef] [PubMed]
  10. Yu CM, Chan JY, Zhang Q, et al. Biventricular pacing in patients with bradycardia and normal ejection fraction. N Engl J Med 2009; 361:2123–2134. [CrossRef] [PubMed]
  11. Choi BJ, Cho KI, Kim SM, Song YJ, Lee HG, Kim TI. Impact of right ventricular apical pacing and its frequency on left atrial function. J Cardiovasc Ultrasound 2012; 20:42–48. [CrossRef] [PubMed]
  12. Sanagala T, Johnston SL, Groot GD, Santucci P, Rhine DK, Varma N. Left atrial mechanical responses to right ventricular pacing in heart failure patients: implications for atrial fibrillation. J Cardiovasc Electrophysiol 2011; 22:866–874. [CrossRef] [PubMed]
  13. Israel CW. The role of pacing mode in the development of atrial fibrillation. Europace 2006; 8:89–95. [CrossRef] [PubMed]
  14. Skanes AC, Krahn AD, Yee R, et al. Progression to chronic atrial fibrillation after pacing: The Canadian Trial of Physiologic Pacing. J Am Coll Cardiol 2001; 38:167–172. [CrossRef] [PubMed]
  15. Healey JS, Toff WD, Lamas GA, et al. Cardiovascular outcomes with atrial-based pacing compared with ventricular pacing: Meta-analysis of randomized trials, using individual patient. Circulation 2006; 114:11–17. [CrossRef] [PubMed]
  16. Sweeney MO, Hellkamp AS, Ellenbogen KA, et al. Adverse effect of ventricular pacing on heart failure and atrial fibrillation among patients with normal baseline QRS duration in a clinical trial of pacemaker therapy for sinus node dysfunction. Circulation 2003; 107:2932–2937. [CrossRef] [PubMed]
  17. Veasey RA, Arya A, Silberbauer J, et al. The relationship between right ventricular pacing and atrial fibrillation burden and disease progression in patients with paroxysmal atrial fibrillation: the long-MinVPACE study. Europace 2011; 13:815–820. [CrossRef] [PubMed]
  18. Papageorgiou P, Monahan K, Boyle NG, et al. Site-dependent intra-atrial conduction delay. Relationship to initiation of atrial fibrillation. Circulation 1996; 94:384–389. [CrossRef] [PubMed]
  19. Yamada T, Fukunami M, Shimonagata T, et al. Effect of atrial septal pacing on P-wave duration dispersion and atrial late potentials in patients with paroxysmal atrial fibrillation. Am J Cardiol 2001; 88:795–798. [CrossRef] [PubMed]
  20. Kristensen L, Nielsen JC, Mortensen PT, et al. Sinus and paced P wave duration and dispersion as predictors of atrial fibrillation after pacemaker implantation in patients with isolated sick sinus syndrome. Pacing Clin Electrophysiol 2004; 27:606–614. [CrossRef] [PubMed]
  21. Bailin SJ, Adler S, Giudici M. Prevention of chronic atrial fibrillation by pacing in the region of Bachmann's bundle: results of a multicenter randomized trial. J Cardiovasc Electrophysiol 2001; 12:912–917. [CrossRef] [PubMed]
  22. Padeletti L, Porciani MC, Michelucci A, et al. Interatrial septum pacing: a new approach to prevent recurrent atrial fibrillation. J Interv Card Electrophysiol 1999; 3:35–43. [CrossRef] [PubMed]
  23. Strohmer B, Pichler M, Froemmel M, Migschitz M, Hintringer F; ELVIS Study Group. Evaluation of atrial conduction time at various sites of right atrial pacing and influence on atrioventricular delay optimization by surface electrocardiography. Pacing Clin Electrophysiol 2004; 27:468–474. [CrossRef] [PubMed]
  24. Szili-Torok T, Bruining N, Scholten M, Kimman GJ, Roelandt J, Jordaens L. Effects of septal pacing on P wave characteristics: the value of three-dimensional echocardiography. Pacing Clin Electrophysiol 2003; 26:253–256. [CrossRef] [PubMed]
  25. Bennett DH. Comparison of the acute effects of pacing the atrial septum, right atrial appendage, coronary sinus os, and the latter two sites simultaneously on the duration of atrial activation. Heart 2000; 84:193–196. [CrossRef] [PubMed]
  26. Hermida JS, Kubala M, Lescure FX, et al. Atrial septal pacing to prevent atrial fibrillation in patients with sinus node dysfunction: results of a randomized controlled study. Am Heart J 2004; 148:312–317. [CrossRef] [PubMed]
  27. Miyazaki H, Noma K, Date T, et al. Atrial septal pacing to resynchronize atrial contraction and improve atrial transport function. J Cardiol 2005; 45:239–246. [PubMed]
  28. Buxton A, Josephson M. The role of P-wave duration as a predictor of post-operative atrial arrhythmias. Chest 1981; 80:68–73. [CrossRef] [PubMed]
  29. Verlato R, Botto GL, Massa R, et al. Efficacy of low interatrial septum and right atrial appendage pacing for prevention of permanent atrial fibrillation in patients with sinus node disease: results from the electrophysiology-guided pacing site selection (EPASS) study. Circ Arrhythm Electrophysiol 2011; 4:844–850. [CrossRef] [PubMed]