Author + information
- Received September 10, 1996
- Revision received December 24, 1996
- Accepted January 20, 1997
- Published online April 1, 1997.
- Atul Prakash, MD, MRCPA,
- Sanjeev Saksena, MD, FACCA,1,*,1,
- Michael Hill, PhDB,
- Ryszard B Krol, MDA,
- Anand N Munsif, MDA,
- Irakli Giorgberidze, MDA,
- Philip Mathew, MSA and
- Rahul Mehra, PhDB
- ↵*Dr. Sanjeev Saksena, Cardiac Medicine and Electrophysiology, 55 Essex Street, Suite 3-2, Millburn, New Jersey 07041.
Objectives. We tested the ability of dual-site right atrial pacing to prevent atrial fibrillation (AF) or atrial flutter induced by eingle-site atrial pacing and correlated its efficacy with clinical patient characteristics, atrial activation times and refractory periods.
Background. Prevention of recurrent AF with long-term dual-site right atrial pacing has been demonstrated in our previous studies. However, the mechanism of antiarrhythmic benefit is unclear.
Methods. Using standard electrophysiologic methods, baseline electrocardiographic and electrophysiologic measurements (mean ± SD) were obtained. Programmed atrial stimulation was performed for AF or atrial flutter induction. Atrial pacing was performed at two drive cycle lengths (600 and 400 ms) and followed by one to three atrial extrastimuli at one to four pacing sites in the right atrium. In patients with inducible AF or atrial flutter, reinduction was then attempted during a dual-site atrial pacing drive train. This was achieved by simultaneously pacing at the high right atrium and coronary sinus ostium at an identical rate to the baseline stimulation, with the atrial extrastimuli being delivered at the pacing site responsible for the initial AF episode initiation.
Results. Twenty patients (10 men, 10 women, mean [±SD] age 64 ± 16 years) with symptomatic AF (n = 10) or atrial flutter (n = 10) were studied. There was a significant abbreviation of the P wave duration to 103 ± 17 ms with dual-site pacing compared with sinus rhythm (120 ± 12 ms, p = 0.005) and high right atrial pacing (121 ± 17 ms, p = 0.005). This was also associated with a characteristic change in P wave configuration with an inferior and leftward axis shift. The effective refractory period at the high right atrium remained unchanged with dual-site atrial pacing compared with single-site high right atrial pacing.
Sixteen patients had inducible AF or atrial flutter and could be tested after dual-site atrial pacing. The induced atrial tachyarrhythmia was suppressed in nine patients (56%), who had either induced AF (n = 5) or atrial flutter (n = 4). The difference in the effective refractory period between the high right atrium and the coronary sinus ostium pacing sites was significantly greater (33 ± 12 ms) in patients with suppression of atrial tachyarrhythmia with dual-site atrial pacing compared with patients without suppression (15 ± 13 ms, p = 0.001). P wave abbreviation did not correlate with arrhythmia suppression. There was no correlation between suppression of inducible AF or atrial flutter and demographic or clinical patient characteristics.
Conclusions. Dual-site right atrial pacing from the high right atrium and coronary sinus ostium can suppress inducible AF or atrial flutter elicited after single-site high right atrial pacing in selected patients. Acute suppression is more likely in patients with greater dispersion of right atrial refractoriness between these two sites.
(J Am Coll Cardiol 1997;29:1007–14)
© 1997 by the American College of Cardiology
Electrical therapies such as single- and dual-site atrial pacing are being evaluated for the long-term prevention of atrial fibrillation (AF). Retrospective and a few prospective studies have suggested decreased frequency of AF recurrences during long-term atrial pacing in patients with sick sinus syndrome ([1–5]). However, the frequency of AF recurrences has been reported to be as high as 19% to 47% in some reports ([1–5]). Potential antiarrhythmic mechanisms for atrial pacing could include suppression of the trigger mechanisms or alteration of the electrical properties of the substrate, or both. Long-term pacing simultaneously from the high right atrium and coronary sinus ostium or the distal coronary sinus has recently been found effective in reducing AF and atrial flutter recurrences in patients with drug-refractory AF ([6, 7]). Loss of dual-site atrial pacing and reversion to single-site pacing has been associated with AF recurrences (). Arrhythmia-free intervals are longer with long-term dual-site right atrial pacing than with high right atrial pacing. The mechanism of the additional beneficial effect of dual-site pacing is unclear. Its electrophysiologic effects as well as the magnitude of the effects has also not been defined. The pattern of electrical activation of the atrium can be expected to be altered by the dual-site pacing mode compared with the single-site atrial pacing mode. In this study, we tested the hypothesis that simultaneous pacing from two atrial sites (high right atrium and coronary sinus ostium) acutely prevents the induction of AF or atrial flutter elicited with atrial extrastimuli during single-site pacing. We also compared the electrophysiologic properties of patients demonstrating AF suppression with those who did not with dual-site atrial pacing.
1.1 Patient selection.
Consecutive patients with a history of AF or atrial flutter with standard indications for electrophysiologic studies were included in the study. Patients were undergoing electrophysiologic studies for evaluation of the arrhythmia mechanism, assessment of their atrioventricular conduction or arrhythmia mapping and ablation. All patients were required to give written informed consent for the electrophysiologic procedure. Only patients with inducible, sustained AF, as measured using a standard programmed stimulation protocol as previously described, were included in this study (unpublished report and ref. ).
1.2 Electrophysiologic studies.
All studies were performed in the electrophysiology laboratory. Patients were premedicated with meperidine or diazepam before the procedure. All antiarrhythmic drugs were discontinued for a minimum of five drug half-lives before the procedure. Standard multipolar catheters with 1-cm interelectrode distance were used for pacing and recording of intracardiac electrograms. These were positioned at the high right atrium, His bundle and right ventricular apex. A fourth catheter was positioned in the coronary sinus with one of the electrodes located at the coronary sinus ostium. Stimulation was performed with rectangular impulses using an EP-2 stimulator (EP Medical Inc.). A standard multichannel recorder was used for hardcopy recording. Three surface electrocardiographic (ECG) leads (I, aVF and V1) were recorded at all times. The intracardiac signals were filtered at a bandpass of 30 to 500 Hz and stored on FM tape. The pacing threshold was determined at the high right atrium and coronary sinus ostium with 2-ms duration pulses. The atrial effective refractory period was determined at twice the diastolic threshold using drive trains of eight stimuli at cycle lengths of 600 and 400 ms. In four patients, sinus rate precluded a drive train of 600 ms, and they were tested at 500 ms. The effective refractory period was measured at the high right atrium using both high right atrial and dual-site pacing drive trains, and also at the coronary sinus ostium using the drive train applied at the coronary sinus site.
Induction of AF was attempted initially from the high right atrium using one to three extrastimuli at drive trains of 600 and 400 ms with coupling interval decrements of 10 ms (unpublished report and ref. ). When adding the second or third extrastimulus (S3or S4, respectively), the previous extrastimulus was fixed at 50 ms above the refractory period. If AF was not induced from the high right atrium, the induction protocol was repeated from the coronary sinus ostium, fossa ovalis and interatrial septum. In patients with inducible, sustained AF or flutter, the entire protocol was repeated using a dual-site atrial drive train. This was achieved by simultaneously pacing from a high atrial site (either the high right atrium, fossa ovalis or interatrial septum—the choice depending on whichever site resulted in successful initiation of the arrhythmia at baseline) along with the coronary sinus ostium; the atrial extrastimuli continued to be applied from the superior atrial site, which was the original site for induction (i.e., high right atrium, fossa ovalis or interatrial septum). Dual-site stimulation was achieved using two electronically coupled pulse generators, resulting in localized bipolar stimulation at both sites. In patients in whom sustained atrial arrhythmia could not be induced using the dual-site drive train, repeat induction from the control site using the single-site drive train was attempted whenever possible. In three patients, this was not feasible because of recurrent spontaneous episodes of AF requiring repeated electrical cardioversion. The coupling interval that induced the arrhythmia was recorded along with the duration and characteristics of the induced arrhythmia.
The effect of dual-site atrial pacing and high right atrial pacing on the atrial effective refractory period at the high right atrium, as well as global atrial activation as represented by the P wave duration and configuration, were analyzed in three scalar ECG plans and compared with sinus rhythm. In addition, electrophysiologic effects (P wave duration, atrial refractoriness, absolute effective refractory period and effective refractory period dispersion between the high right atrium and coronary sinus ostium) and clinical variables (age, gender, heart disease status, left atrial diameter and left ventricular ejection fraction) were compared in patients demonstrating induced AF suppression and those in whom AF could not be suppressed using the dual-site drive train.
1.3 Definitions and end points.
The following ECG and electrophysiologic definitions were used in this study: 1) atrial fibrillation = irregular atrial activation on the surface ECG at cycle length ≤180 ms with marked beat to beat variation in atrial activation configuration and cycle length; 2) atrial flutter = regular atrial activation on the surface ECG with a cycle length 180 to 250 ms and the presence of discrete flutter waves; 3) atrial tachycardia = regular atrial activation on the surface ECG with a cycle length >250 ms; 4) nonsustained arrhythmia = atrial tachyarrhythmia <30 s in duration; 5) sustained arrhythmia = atrial tachyarrhythmia ≥30 s in duration; 6) suppression of induced arrhythmia with dual-site pacing = inability to induce sustained arrhythmia using a dual-site drive train and high right atrial extrastimuli after completion of the entire induction protocol; 7) effective refractory period dispersion between the high right atrium and coronary sinus ostium = maximal effective refractory period difference between the two sites at either of the two drive cycle lengths (600 and 400 ms); and 8) P wave duration = maximal duration as measured in surface ECG leads I, aVF and V1.
1.4 Analysis and statistical methods.
The end point was 1) completion of the entire stimulation protocol using the dual-site drive train in all patients with inducible AF with single-site pacing, or 2) induction of AF during the dual-site stimulation protocol. Continuous variables were expressed as mean values ± SD. The association of patient characteristics and electrophysiologic measurements with the suppression of AF was studied using a multivariate logistic regression model. All tests of significance were based on the likelihood ratio statistic. Furthermore, continuous measures such as P wave duration between single- and dual-site pacing were analyzed using the paired and unpaired Student ttest for paired and unpaired data, respectively.
2.1 Patient group (Table 1).
Twenty patients (10 men, 10 women, mean age 64 ± 16 years) were included in the study. Fifteen patients had evidence of structural heart disease. The mean left atrial diameter, as determined on echocardiography, was 39 ± 7 mm, and the mean left ventricular ejection fraction was 44 ± 15%. Ten patients had paroxysmal atrial flutter and 10 patients had AF as their clinical arrhythmia.
2.2 Electrophysiologic effects of dual- and single-site atrial pacing.
2.2.1 Effects on global atrial conduction and activation.
The mean P wave duration in sinus rhythm (120 ± 12 ms) was not significantly different from that during high right atrial pacing (121 ± 17 ms). The mean P wave duration was, however, significantly shortened (103 ± 17 ms) with dual-site atrial pacing (p = 0.005; Fig. 1). The configuration of the P wave was also significantly altered with dual-site atrial pacing compared with sinus rhythm and high right atrial pacing. The characteristic P wave configuration during dual-site right atrial pacing was an initial positive deflection followed by a terminal negative component in lead II of a 12-lead ECG. There was a significant inferior and leftward shift of the P wave axis with dual-site atrial pacing compared with sinus rhythm and high right atrial pacing.
2.2.2 Effects on atrial refractoriness (Fig. 2).
Table 1details the individual values of the atrial effective refractory period at the high right atrium and coronary sinus ostium for all patients. These values could not be measured in four patients at a drive cycle length of 600 ms. Mean values are shown with (lower figure) and without (upper figure) these patients. The mean effective refractory period at the high right atrium could be measured at a drive cycle length of 600 ms in 15 patients. It was 232 ± 29 ms compared with 218 ± 23 ms at a drive cycle length of 400 ms (p = 0.15). The mean effective refractory period at the coronary sinus ostium was 247 ± 29 ms at a drive cycle length of 600 ms compared with 232 ± 26 ms at a drive cycle length of 400 ms (p = 0.2). There was no significant difference between the mean effective refractory period at the high right atrium and coronary sinus ostium at both drive cycle lengths of 600 and 400 ms in the entire sample (Fig. 2). The mean effective refractory period at the high right atrium using the dual-site pacing drive was 228 ± 29 ms. This was not significantly different from the mean effective refractory period at the high right atrium using the high right atrial pacing drive (p > 0.2) and permitted similar coupling intervals for atrial extrastimuli during single- and dual-site pacing drives.
2.3 Effects of dual-site pacing on induced arrhythmia.
Sixteen of the 20 patients had inducible, sustained AF (n = 9) or atrial flutter (n = 7) and could be tested for suppression with dual-site atrial pacing. Using dual-site atrial pacing the inducible, sustained atrial arrhythmia could be suppressed in nine patients for an overall efficacy of 56%. Five of the suppressed patients had AF and four had atrial flutter as the initial induced arrhythmia. Thus, the efficacy of suppression of the induced arrhythmia was 55% for AF and 57% for atrial flutter (Fig. 3). Table 2depicts the induced arrhythmia, mode of induction and results of dual-site atrial pacing in each of the 16 patients with inducible AF or atrial flutter.
Fig. 4shows the induction of AF in a patient, using the single-site drive train with a single extrastimulus delivered from the high right atrium. With the dual-site drive train, AF is not induced with one, two or three extrastimuli from the high right atrium. Fig. 4shows the use of two extrastimuli that failed to initiate AF; however, it is reinduced when the single-site drive train and a single extrastimulus were used. Six of nine patients who had demonstrated suppression with dual-site pacing had reinitiation of AF attempted for a second time in the single-site mode. All six patients had AF reinduced.
2.4 Comparison of clinical and electrophysiologic characteristics of patients with inducible and suppressed arrhythmia.
There was no correlation between suppression of AF with dual-site atrial pacing and age, gender, heart disease status or the clinical/induced arrhythmia in the logistic regression model. A smaller left atrial diameter (37 ± 5 mm) was observed in patients with AF suppression compared with patients with inducible AF (42.5 ± 8 mm), but this was not significantly different (p > 0.2). Patients with AF suppression had a higher mean left ventricular ejection fraction (46 ± 13%) compared with patients with inducible AF (42 ± 15%). This difference, however, was also not significantly different (p > 0.2).
The mean P wave duration of patients with AF or atrial flutter suppression during dual-site right atrial pacing was 119 ± 21 ms. This was not significantly different from the mean P wave duration of patients with continued inducible arrhythmia during this pacing mode (122 ± 23 ms, p > 0.2). The mean reduction seen in P wave duration with dual-site atrial pacing (17 ± 13 ms) compared with high right atrial pacing in patients with arrhythmia suppression was also not significantly different when compared with this value in patients who continued to have inducible AF (21 ± 17 ms, p > 0.2).
The mean individual patient difference in the effective refractory period between the high right atrial and coronary sinus ostial sites (mean dispersion of atrial refractoriness) was 33 ± 12 ms in patients with suppression of induced AF during dual-site right atrial pacing. Mean dispersion of atrial refractoriness was, however, significantly less in patients with continued inducible AF during dual-site right atrial pacing (15 ± 13 ms, p < 0.001; Fig. 5). In the logistic regression model, this was also highly significantly correlated with AF suppression (p = 0.0007). The mean effective refractory period at the high right atrium and coronary sinus ostium in both groups of patients was similar, however.
This study suggests that dual-site atrial pacing from the high right atrium and coronary sinus ostium can reduce global atrial activation time and can acutely prevent the induction of both AF and atrial flutter elicited with single-site right atrial pacing. The electrophysiologic basis for AF prevention with dual-site atrial pacing can be considered based on existing information on AF initiation and maintenance. Abnormalities of atrial refractoriness and conduction have both been described in patients with long-standing AF.
3.1 Atrial refractoriness.
A short effective refractory period at the high right atrium has been described in patients with AF. In one series, the mean effective refractory period in patients with AF was 206 ± 24 ms compared with 233 ± 28 ms in patients without AF (p < 0.02) (). Similar observations have been noted by other investigators, with the mean effective refractory period in the AF population being 216 ms (). Wijffels et al. (), in a goat model of induced AF, noted that the effective refractory period in the atrium became progressively shortened with increasing duration of the arrhythmia. This has been described as atrial electrical “remodeling.” This remodeling may account for the abbreviated effective refractory periods.
Similarly, both experimental and clinical studies have observed that the normal rate adaptation of the atrial effective refractory period at shorter drive cycle lengths is attenuated and can even be reversed in some animals and patients with AF ([10–12]). The mean effective refractory period at the high right atrium in our patients was somewhat longer compared with that reported in published reports. We did not observe a significant shortening of the mean effective refractory period at both the high right atrium and coronary sinus ostium using a shorter basic drive. The use of dual-site atrial pacing did not affect the mean effective refractory period at the high right atrium. This would suggest that AF prevention by this mode of pacing was not related to a change in the atrial refractoriness at the pacing site, nor did this pacing mode interfere with atrial extrastimuli achieving comparably close coupled intervals as with single-site pacing.
An abnormal dispersion of atrial refractoriness in patients with AF has been noted by many investigators ([13–15]). Josephson et al. () observed a mean difference of 35 ms between the effective refractory period at the high right atrial and coronary sinus ostial regions in patients with AF, although this difference was only 5 ms in the control group.
Papageorgiou et al. () recently reported on atrial refractoriness in patients without a history of atrial arrhythmia or heart disease. In this group, coronary sinus effective refractory periods were longer than high right atrial effective refractory periods by 32 ms. In our patients with symptomatic arrhythmia and heart disease, this difference was 15 to 18 ms depending on the pacing rate. Eight of 17 patients in the Papageorgiou study had inducible AF. Interestingly, the difference in the mean effective refractory period between the coronary sinus and high right atrium was higher in the group without inducible AF at baseline. However, in our patients with spontaneous AF, this mean difference at baseline was 17 ms in patients with inducible arrhythmia and 5 ms in patients without inducible arrhythmia. There was a significant difference based on continued AF induction after dual-site atrial pacing in the 16 patients with induced arrhythmia at baseline. In patients in whom dual-site atrial pacing suppressed inducible AF, this difference in the mean effective refractory period between the two sites (33 ± 12 ms) was significantly greater than that in patients in whom dual-site atrial pacing was not successful in suppression of AF (15 ± 13 ms). Dual-site atrial pacing could conceivably lead to a reduction of atrial dispersion of refractoriness and, more importantly, a change in recovery of atrial excitability. This earlier recovery of excitability with elimination of a potential zone of “slow” conduction could be an important mechanism of AF prevention with this pacing mode.
3.2 Intraatrial conduction.
Abnormalities of atrial conduction have been described in relation to AF. Significantly prolonged P wave duration reflecting global conduction delay has been described by many investigators ([17, 18]). There is also an alteration of the activation sequence of the atria, as reflected by a change in P wave configuration. We observed a modest global conduction delay in our patients (mean P wave duration 120 ± 12 ms). Regional intraatrial conduction time, as measured by activation times from the high right atrium to the His bundle and high right atrium to the coronary sinus region, is prolonged in patients with AF (). This interval is significantly abbreviated by dual-site atrial pacing owing to two simultaneous atrial activation wave fronts and manifests as abbreviation of the P wave (). Reduction in P wave duration with dual-site and biatrial pacing has been extensively documented in our previous studies and in work from other centers ([6, 7, 17, 18]). This is in contrast to biventricular pacing where the higher conduction velocity and the specialized conduction system provide rapid activation of right and left ventricle. However, even during biventricular pacing, patients with left bundle branch block also show abbreviation in QRS duration.
The abbreviation in P wave duration with dual-site pacing, however, did not directly correlate with AF suppression or continued reinduction in this study. This has two possible explanations. One potential explanation is that an improvement in global atrial conduction may be important only if the baseline global conduction delay was greater in magnitude, as occurs in patients with type 2 atrial flutter and advanced intraatrial block (). This may be particularly relevant for arrhythmia maintenance in large macroreentrant circuits such as type 2 atrial flutter. However, there also may be critical sites for AF initiation that may be affected even by the modest changes in atrial activation time seen in this study. Thus, a second mechanism may be that a change in regional atrial activation pattern may be more important for prevention of AF onset. Papageorgiou et al. () have suggested that nonuniform anisotropic properties exist in the posterior triangle of Koch. We have recently reported that AF initiation is commonly associated with the appearance of intraatrial conduction delay for the initiating extrastimulus at the septal and coronary sinus ostial regions and much more infrequently at the crista terminalis (). Pacing at the coronary sinus ostium or using dual wave fronts encompassing this region may modify local activation and also improve recovery of excitability, eliminating conduction delays that may be critical in AF onset.
3.3 Mapping atrial activation during dual-site atrial pacing in patients with AF.
Catheter and intraoperative atrial mapping has demonstrated that dual-site atrial pacing advances activation in most of the right atria, including the crista terminalis and interatrial septum and much of the left atrium, including the roof and the left atrial appendage ([18, 19]). We have noted that a change in activation pattern also alters recovery of excitability for closely coupled atrial extrastimuli. This pacing pattern can also change the coupling interval for atrial premature beats that trigger AF and avoid conduction delay in more distant atrial sites for such beats. This can alter the critical conditions for reentry in the initiation of AF. For example, high right atrial extrastimuli fail to achieve comparable coupling intervals when delivered after coronary sinus pacing drive trains compared with high right atrial pacing alone ([21, 22]). This can reduce the window for AF induction by atrial extrastimuli (). Recent mapping data in induced AF in our laboratory indeed suggest that in a majority of patients, the sites of initial conduction delay for the first reentrant beat during AF appear to reside in the low interatrial septum and the proximal to midcoronary sinus or adjoining sites in the left atrium (). Papageorgiou et al. () recently confirmed prolongation of intraatrial conduction times to the posterior triangle of Koch and electrogram widening at this site. These very sites are pre-excited by dual-site right atrial pacing. Furthermore, recent electrogram analysis in type 2 atrial flutter (excluding right atrial flutter) with a monomorphic or polymorphic configuration has been insightful (). This arrhythmia is usually the prelude to sustained AF. Chaotic or fragmented electrical activity suggesting sites of delayed conduction was most frequently seen in the coronary sinus or adjoining left atrial and interatrial septal sites. Although it is unlikely that the electrophysiologic effect of overdrive dual-site pacing would persist beyond the first few beats of AF onset acutely, remodeling of these sites in the low right atrium, coronary sinus and adjoining left atrium is conceivable during long-term pacing, potentially enhancing efficacy further.
3.4 Study limitations.
One potential limitation of this study is that the effects of dual-site right atrial pacing on arrhythmia induction could not be assessed in the four patients without inducible AF. These individuals may or may not have shown similar favorable effects. It is important to recognize the pilot nature of this investigation with the need for continued study in a larger study group. Although the initiation of AF by programmed stimulation has been validated in our laboratory, reproducibility of this approach is under continued study. All six patients with AF suppression using dual-site pacing had reproducibly induced AF. In our initial studies, the programmed atrial stimulation protocol used in this study had a sensitivity of 89% and a specificity of 95% in patients with and without atrial flutter and AF (). In a meta-analysis of the published data, sensitivity of similar protocols using only one atrial site ranged from 72% to 81% ([24, 25]). The higher sensitivity in our study can be explained on the basis of using more atrial sites. The reproducibility of a particular extrastimulus level has not been widely studied, but we completed the entire protocol in patients before concluding that AF initiation was suppressed. Our study design applied dual-site pacing after single-site pacing, and this could potentially introduce bias. A randomized, prospective study design may address this concern. The role of autonomic nervous system activation during AF initiation and cardioversion in some patients is also unclear with regard to its influence on subsequent dual-site pacing. Autonomic blockade may be considered in future studies.
The main effects of dual-site atrial pacing appear to be an abbreviation of global atrial conduction, a change in the atrial activation pattern and, probably, a reduction in right atrial dispersion of refractoriness. We conclude that dual-site right atrial pacing can successfully acutely suppress the induction of both AF and atrial flutter in selected patients. This effect is probably related to advancing activation of large regions in both atria, improving global atrial conduction and reducing the dispersion of atrial refractoriness, resulting in earlier recovery of excitability. This recovery of excitability may be valuable in reducing or modifying “slow” conduction seen with atrial premature beats that trigger AF. Further mapping studies are presently being carried out to analyze the effect of these pacing modes on regional atrial activation.
↵1 Dr. Saksena has been a consultant to Medtronic and other device manufacturers.
- atrial fibrillation
- electrocardiogram, electrocardiographic
- Received September 10, 1996.
- Revision received December 24, 1996.
- Accepted January 20, 1997.
- The American College of Cardiology
- Sgarbossa EB,
- Pinski SL,
- Maloney JD,
- et al.
- Saksena S,
- Prakash A,
- Hill M,
- et al.
- Daubert C,
- Mabo PH,
- Berder V,
- Gras D,
- Leclercq C
- Krol RB,
- Saksena S,
- Giorgberidze I,
- Prakash A
- Cosio FG,
- Palacios J,
- Vidal JM,
- et al.
- Josephson ME
- Wijffels MCEF,
- Kirchhof CJHJ,
- Dorland R,
- Allessie MA
- Misier AR,
- Opthof T,
- Van Hemel N,
- et al.
- Papageorgiou P,
- Monahan K,
- Boyle NG,
- et al.
- Prakash A,
- Saksena S,
- Kaushik R,
- Krol RB,
- Munsif AN,
- Mathew P
- Munsif AN,
- Prakash A,
- Krol RB,
- Mathew P,
- Lewis C,
- Saksena S
- Saksena S,
- Giorgberidze I,
- Prakash A,
- et al.
- Prakash A,
- Hill M,
- Giorgberidze I,
- et al.
- Hill MRS,
- Mongeon LR,
- Mehra R
- Giorgberidze I,
- Munsif AN,
- Krol R,
- Mathew P,
- Prakash A,
- Saksena S