ATRIAL FIBRILLATION
Impact of anti-tachycardia pacing on atrial fibrillation burden when added on top of preventive pacing algorithms: results of the prevention or termination (POT) trial
1 Thorax Institute, Hospital Clinic, University of Barcelona, Institut d'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Villarroel 170, Barcelona 08036, Catalonia, Spain; 2 Arrhythmia Unit, Cardiology Department, Hospital Clinico Universitario, Valencia, Spain; 3 Arrhythmia Unit, Cardiology Department, Hospital General Alicante, Alicante, Spain; 4 Arrhythmia Unit, Cardiology Department, Hospital de Navarra, Pamplona, Spain; 5 Arrhythmia Unit, Cardiology Department, Hospital de León, Spain; 6 Universitat Politècnica de Catalunya, Barcelona, Spain; 7 Scientific and Clinical Department, Medtronic Ibérica, Barcelona, Spain
* Corresponding author. Tel: +34 93 2275551; fax: +34 93 4513045.E-mail address: lmont{at}clinic.ub.es
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionFundingAppendixReferences |
|---|
Aims: The efficacy of preventive pacing algorithms (PPA) and anti-tachycardia pacing (ATP) in reducing atrial fibrillation (AF) burden remains controversial. The aim of this study was to assess whether ATP on top of PPA decreases AF burden.
Methods and results: A series of 199 consecutive patients, with conventional indications for pacing, and documented AF, received a DDDR (rate adaptive dual chamber pacemaker) pacemaker with ATP capabilities (AT 500 Medtronic). After 3 months of conventional DDDR pacing at 70 b.p.m., AF burden was analysed. If patients had >30 min/week of AF, they were randomized to PPA or to PPA+ATP for 3 months (period 1). They were then crossed to the alternative therapy (period 2) and followed three additional months with a 1-month wash out period in-between. A group of 85 patients were randomized. Mean age 68 ± 8 years, 61% men. Both groups showed a significant decrease in AF burden at the end of period 1 (64 and 81%, respectively).
Conclusions: Atrial pacing with PPA decreases AF burden in patients with pacing indication. We did not observe a further decrease in AF burden or in the number of episodes when adding ATP on top of PPA.
Key Words: Anti-tachycardia pacing, ATP, Preventive pacing algorithms, Atrial fibrillation burden, POT trial, AT500
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAppendixReferences |
|---|
Atrial pacing has been shown to decrease the incidence of atrial fibrillation (AF) in patients with sick sinus disease as compared to VVI mode.1
This observations along with the acquired knowledge of the underlying mechanisms of AF2 prompted the development of antifibrillatory algorithms to prevent atrial tachycardia (AT) and AF. The efficacy of these antifibrillatory algorithms is still under debate. Although some studies have reported a benefit in reducing AF burden;3,4 the majority have not found clinically relevant improvement.5–8 The ATP algorithms initially introduced in pacemakers and implantable cardioverter defibrillators (ICDs) were designed to treat AT/AF episodes in order to attempt to decrease AF burden, however, they have not been evaluated as to their ability to prevent AT or AF. Analysis of stored electrograms has shown that many AT/AF episodes are regular at the beginning of the episode and therefore, may respond to ATP. Although several studies have shown the efficacy of ATP in terminating AT/AF episodes, the effect in decreasing the arrhythmia burden has not been fully tested and it is still under debate. The objective of this study was to analyse whether the addition of ATP to preventive algorithms may result in a decrease of arrhythmic burden.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAppendixReferences |
|---|
Study population
Patients enrolled in the trial were eligible, if they had a class I or class IIa indication for dual chamber pacemaker9
and had documented atrial tachyarrhythmias (
2 episodes lasting more than 30 s in the previous 6 months). Patients in permanent AF and those with a contraindication for chronic anticoagulation were excluded.
Device characteristics and programming
All patients received an AT500TM (Medtronic Inc, Minneapolis, MN, USA) dual-chamber rate responsive pacemaker with Prevention Pacing Algorithms (PPA), ATP therapies and the capability to measure AF burden. Its precision and measurements have been previously described.7,10 The PPA algorithms include: (i) Atrial Preference Pacing (APP), designed to maintain the pacing rate just above the intrinsic rate; (ii) Atrial Rate Stabilization (ARS), designed to avoid short–long intervals following a premature atrial complex; and (iii) post-mode switch overdrive pacing (PMOP), designed to inhibit the early recurrence of atrial tachyarrhythmia following termination of atrial arrhythmias by transiently pacing at a higher rate. Available ATP therapies include: (i) Burst+ (constant drive of pulses followed by up to two extra-stimuli); and (ii) Ramp (decrementing drive of pulses). The ATP is delivered if any AT episode persists longer than a programmable ‘time to first therapy’,11 which in our study was set to 0 min (immediately after detection). All pacemaker settings are shown in Table 1.
|
After implant, the device was programmed to DDD/DDDR (rate adaptive dual chamber pacemaker) pacing mode with a lower rate between 60 and 70 b.p.m. Atrial rhythms detected with a cycle length from 180 to 400 ms were classified as AT, and those detected with a cycle length between 100 and 170 ms, were classified as AT or AF depending on R–R irregularity.10
,12 This programming was selected in order to detect as many episodes as possible and to facilitate device ATP treatment. The number of episodes and burden was based on device memory and counters.
Study design
The Prevention Or Terminiation (POT) trial is a multi-centre, single-blind, prospective, randomized, crossover study. The study was approved by the local ethics committee of each participating institution and all patients provided written informed consent prior to enrollment. Baseline medical history and physical examination were performed before pacemaker implantation. Patients first underwent a 1 month of antiarrhythmic medication (AAD) stabilization followed by a 3-month run-in period, with ATP or PPA therapies switched off (detection on). Patient who, during this time, experienced 3-months of cumulated AF burden over 30 min/week as detected by the device, were randomized to Group A or Group B. This cut-off point was decided by the Steering Committee at the design phase of the study. The randomization scheme was developed by an external statistician using a random number generator to select random permuted blocks of 4. Patients did not have knowledge of their randomization assignment. Patients were randomized to Group A (PPA first, then ATP+PPA) or Group B (ATP+PPA first, then PPA) (Figure 1). The first month after crossover was considered a wash out period and any AF recorded during this period was not used for the primary analysis.
|
Follow-up visits were performed at baseline, and 4-months, 7-months and 11-months post-enrolment, which included physical examination, spontaneous AT/AF episodes registration, and pacemaker check up. In patients with more than 20 episodes of AT recorded by the device at any visit, the collection of the save to disk (device memory) was performed on a monthly basis to ensure as many EGM stored episodes as possible, since the device stores a maximum of 35 episodes with EGM. Anyway, the total number, and episode description are stored in the memory device, even if EGM are not available. During this intermediate visits no change in programmed parameters were allowed.
Data analysis
Analysis was performed using the intention to treat principle (ITT). Last value carried forward was imputed to loss of follow-up cases. The following assumptions were made to account for death or therapy failure. During the run-in period, a patient who required cardioversion (CV) was considered to have met the AF burden requirements for randomization. During the crossover phase, if a patient required two or more CV in the same period, it was considered therapy failure. In both cases, as well as any death, the maximum possible AF burden per week was 10 080 min (7 days x 24 h x 60 min). There were two periods where CV were allowed and not penalized: (i) during the first month of drug stabilization (from implant to randomization) and (ii) during the wash-out month (from crossover month 7 to month 8). Secondary variables, such as number and duration of episodes, are presented without imputed values for death and treatment failures.
A logarithmic transformation13 was employed to improve the fit of the skewed distribution of the AF burden to the normal model. Crossover was analysed by the standard t-test and results are presented with standard deviations for baseline data and 95% confidence intervals (CI) for outcome effects. Data are expressed as burden in %, mean and median and for comparison the geometric mean was used.
The rationale to use a crossover design was based on 3 assumptions: (i) a 1-month wash-out period was considered sufficient to remove delayed or residual effects of the first period programming; (ii) the patients had enough stability in order to have the conditions to study the treatment effects at the same circumstances on both periods;14 (iii) intra-patient burden variability was small enough to justify a logistically feasible sample size.
For sample size calculations, a reduction in AF burden of 22% was considered reasonable, implying a log difference mean of 0.2, with a standard deviation of 0.2 and a CI amplitude from 16.11 to 28.48%. This resulted in a minimum sample size requirement of 60 patients who met randomization criteria (30 per group). To account for infrequent protocol deviations, sample size was fixed at 70 patients.
Statistical significance and confidence levels were set at the standard values of 0.05 and 0.95.
During the study, the number of patients meeting the AF burden requirement (>30 min/week) for randomization was less than anticipated, therefore, we extended the inclusion period, which led to the enrolment of 199 patients.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAppendixReferences |
|---|
Patient enrolment occurred between March 2001 and April 2004, and the study ended in June 2005. From 199 screened patients, 85 (43%) patients (mean age 68 ± 8 years) fulfilled the selection criteria of AF burden of at least 30 min/week and were randomized.
During follow-up, 1 patient died and 2 patients were considered as treatment failures (1 patient with recurrent atrial flutter requiring ATP, and one patient with symptoms due to the ATP treatment) (Figure 2). These three patients were assigned the maximum possible burden (10 080 min/week) during the follow-up time (3 months). After 7 months, five patients were lost to follow-up. In 3 patients, crossover was not performed as expected, so they were maintained in their initial treatment settings according to randomization schedule and ITT. We also included, in the ITT analysis, six (7%) patients that were randomized but at the device data monitoring review, it was determined they did not meet the AF burden cut-off criterion after the run-in period. We also repeated the analysis without including these patients and they did not change significantly.
|
Clinical characteristics of patients allocated to both groups were similar (Table 2). Descriptive results of AF burden, number and length of episodes during run-in, first and second period for both groups are shown in Table 3. A total of 60 and 51 patients had more than 20 AF/AT episodes during the 3 months run-in period and period 1, respectively. The percentage of atrial and ventricular pacing during run-in, first and second period for both groups are shown in Table 4.
|
|
|
The majority of patients (63%) were receiving AAD during the study (27% class I, 27% amiodarone and 45% beta-blockers plus class I). AAD was kept stable during the study, after the first month run-in period.
In all patients, regardless of group assignment, there was a significant decrease in AF burden at the end of period 1 from run-in period (Figure 3). The first period had 28% (95% CI, 15–47%) of the initial burden, equivalent to 72% decrease. At the end of period 2, the AF burden was 57% from period 1 burden (95% CI, 29–111%) that was not statistically significant.
|
Regarding treatment effects, PPA had a non-significant difference in AF burden reduction (Figure 4) as compared to PPA+ATP (90% of the AF burden when treated only with PPA relative to PPA+ATP, which means 10% of AF burden reduction with only PPA; 95% CI: 46–178%). Similar results were observed for number and duration of episodes: although not significant, a better point estimates were obtained with the PPA treatment only. The reduction in number of episodes among patients treated only with PPA was 37%; 63% of the number of episodes when treated only with PPA relative to PPA+ATP (95% CI, 39–102%). The reduction on episode duration among PPA only patients was 8% [92% of the episode duration when treated only with PPA relative to PPA+ATP (95% CI, 46–183%)]. The delayed, crossover or residual effect of the first treatment over the second period was not significant.
|
Mean global efficacy of ATP estimated by the device was 65%. We analysed the electrograms in a subgroup of 30 patients, in whom the termination of episodes with ATP was achieved in 57% (35–93%) of the cases. Efficacy was lower (35%) when the electrograms were classified as irregular and continuous electrical activity and higher (61%) in highly organized electrograms.
A total of 32 CV were performed in 22 patients during the study. During the drug stabilization period, four CV were performed on four patients. During the run-in period (therapies OFF, detection ON), 15 CV (three in Group A and 12 in Group B) were performed. In the first randomization period seven CV were performed on five patients (three in Group A and four in Group B), 2/7 patients received two CV each and were assigned the maximum burden (one patient in Group A and one in Group B). During the wash-out period between crossover assignments, three CVs were performed on three patients. Finally, during the second crossover period, three CV were performed in two patients (two CV in the same patient in Group A, who was assigned the maximum burden). In addition, two patients required AV node ablation during the study (one patient in Group A during the first period, and one in Group B in the second period).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAppendixReferences |
|---|
The role of atrial pacing in the prevention of AF remains controversial. The initial study by Andersen et al.1
in patients with sick sinus disease, who required pacing, showed that atrial pacing decreased the incidence of AF as compared to VVI pacing. The MOST and CTOP studies showed that pacing in DDDR mode reduced the incidence of AF compared to VVI.15,16 These encouraging initial results with atrial pacing prompted the introduction of using preventive pacing algorithms (PPA) along with ATP capabilities in an attempt to reduce AF in various populations. So far, a few studies have been able to report positive results from using such algorithms in the POT of AF,2,3 while others have failed to demonstrate any significant benefit.5–8 The therapy has been applied to the following populations: patients receiving ICD,6,17,18 patients with paroxysmal AF and indication for AV node ablation19 or patients with pacing indication.3,13,20 Furthermore, the preventive algorithms and ATP have been applied altogether in some studies, making it impossible to delineate the possible benefit of each individually. So far, the use of pacemakers to prevent AF in patients who do not meet pacemaker indications has only been performed in research studies and in highly select populations.
To our knowledge, the POT trial is the first study to evaluate the potential benefit of ATP in reducing AF burden when applied on top of preventive algorithms in patients with high AF burden. Our results show a significant reduction of AF burden in both therapy arms but without significant differences between groups. The before and after differences are consistent with a positive evolution since the moment pacing was started. This kind of progressive reduction in AF burden in both therapy arms has also been observed in other studies even in the DDDR pacing without preventive algorithms.3 Therefore, it seems that in patients with sick sinus disease and AF, there is a significant improvement in AF burden with regular atrial pacing, and the additional benefit offered by preventive algorithm is difficult to uncover. Lee et al.7 of the ATTEST study demonstrated that the AT 500 (Medtronic) had an accurate detection scheme and provided ATP therapy with 54% efficacy, but failed to demonstrate any significant AF burden reduction in patients with and without activated algorithms and pacing. Different from our study, ATTEST compared the efficacy of PPA and ATP together vs. conventional pacing, but did not evaluate the potential additional benefit of applying ATP on top of PPA. Furthermore, there was no predefined minimum AF burden requirement. Given the wide dispersion of AF burden normally observed, it may be desirable to try to homogenize the population before randomization in any study designed to evaluate the results of a therapy.
This crossover study was not able to demonstrate any effect on the AF burden, with the addition of ATP to PPA in a series of pacemaker-indicated patients with high AF burden. When this study began in 2001, many ongoing studies evaluating the reduction of AF with APP and/or ATP were using AF burden as an endpoint. Today, we know that using this endpoint alone may have limitations to be able to demonstrate clinical relevance.21–23
The 28% decrease in AF burden observed in both groups that occurred between the run-in and the first crossover period could be due to the ‘regression to the mean’ phenomenon. Since we saw high intra-patient variability, and since patients were selected only after they had experienced a at least 30 min of AF burden per week, regression to the mean can explain an evolution up to 45% (55% decrease) of the run-in burden.23 However, regression to the mean may only explain a reduction from the screening to the first period, but not from the first to the second period because the cut-off criteria were applied at the end of the screening period and because the first period had no further selection of patient selection criteria. Furthermore, the improvement observed over time in both groups, does not interfere with the main conclusion that we did not observed any incremental benefit of ATP on top of PPA.
A possible limitation is that the device was programmed to the DDD/DDDR mode. The AV interval programming was left to the discretion of the investigator; therefore, if intrinsic conduction was not recognized, a significant percentage of ventricular pacing could have occurred, having a negative effect on AF burden. A strategy to minimize right ventricular stimulation could help prevent this possible effect.24
The study was designed to provide short CIs for the intervention effect. Unfortunately the variability of the AF burden was much higher than expected resulting in a study with wide CIs and reduced power. We estimated the between-patient standard deviation in the log scale to be 3.46, showing a high between patient variability. Although the intra-patient variability was much smaller (1.16), it is still large compared to some desirable treatment effects. For example, a treatment effect reducing the AF burden to 80% will have in future crossover designs a power of 24%, clearly insufficient, but much better than the corresponding 6% power of a parallel design with 85 patients in each treatment arm (Figure 5). As another example, the crossover design achieve the standard 80% power to test a treatment effect of 61% AF burden, but the parallel design would had only 14% power. So, although the crossover design controls the higher between-patient variability, it remains a highly uncontrolled source of variability that lowers precision and power. Those results are in accordance to the conclusion suggested by Padeletti et al.21 and Botto et al.22 in the sense that this variability implies that larger number of patients and longer observation periods are needed in order to highlight possible treatment effects. This could be a limitation of using the AF burden as a study variable.
|
We also noticed there was a progressive, significant, decrease of the AF burden that could be due to a cumulative benefit from conventional DDDR pacing rather than the addition of PPA or PPA+ATP. This finding suggests that perhaps a crossover design was not ideal to study this unstable condition. Although logarithms are useful to analyse the relative reduction instead of the absolute reduction, it is not the a desired method if two conditions exist: (i) if the treatment effect may vary upon the magnitude of the baseline AF burden and (ii) if the reduction from 60 to 30 is not the same as the reduction from 2.4 to 1.2 as suggested by Lewalter and Luderitz.25
These effects were unexpected at the time of study design. Another limitation of the study is the lack of information about quality of life. Although AF burden may be an objective endpoint, a demonstration of improvement in quality of life is considered nowadays a very important issue.
Finally, previous studies have found a positive effect of preventive algorithms in selected patients4,26 suggesting that certain characteristics may identify potential responders. Although we found patients with an extremely positive response to PPA+ATP, we have not specifically analysed the characteristics of these patients; therefore, PPA+ATP could be of potential benefit in an specific subgroup of patients.
To sum up, we consider our results valuable, since they showed us that: (i) there is an overall positive trend in the reduction of AF burden, as shown by the statistical analysis and the analysis of individual responses; (ii) AF burden is a highly variable outcome requiring a larger sample size in order to consider it as a principal response variable in a clinical trial; and (iii) the crossover design should be carefully considered in studies using AF burden as an endpoint.
|
Funding |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAppendixReferences |
|---|
The study was supported by a grant from Medtronic Ibérica, S.A. Funding to pay the Open Access publication charges for this article was provided by Fundació Clínic per a la Recerca Biomédica.
Conflict of interest: Dr L.M. has conducted research and given lectures for Medtronic, Guidant & St Jude Medical. M.R. and X.N. are employed by Medtronic.
|
Appendix |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAppendixReferences |
|---|
POT trial group
POT Participating Investigators, Centres, and Enrolments: Dr L. Mont, H.Clínic, Barcelona (78); Dr J. Gabriel-Martínez, H.General, Alicante (18); Dr R. Ruiz-Granell, H.Clínico, Valencia (13); Dr Bertomeu, H.San Juan, Alicante (12); Dr Delclós, H.del Mar, Barcelona (11); Dra.M. Fidalgo, H.León, León (10); Dr R. Carmona, H.Navarra, Pamplona (10); Dr Marrero, H.Univ. Canarias, La Laguna (10); Dr J.Sotillo, H.Arnau de Vilanova, Valencia (9); Dr J. Balaguer, H.General, Guadalajara (9); Dr X. Beiras, H.Xeral de Vigo, Vigo (7); Dr G. Fdez. Mora, H.Inf.Cristina, Badajoz (5); Dr A. Grande, H.Severo Ochoa, Leganés (4); Dr Benezet, H.Alarcos, Ciudad Real (2); and Dr Mnez. Sande, Com.H.Santiago, S.Compostela (1).
POT Steering Committee: Dr L. Mont, H.Clínic, Barcelona; Dr R. Ruiz-Granell, H.Clínico, Valencia; Dr J. Gabriel-Martínez, H.General, Alicante; Dra. M. Fidalgo, H.León, León; Dr R. Carmona, H.Navarra, Pamplona; and Dr X. Navarro, Medtronic Ibérica, Madrid.
Study Management: Mireia Riera, Medtronic Ibérica, Spain and Statistical Analysis: Erik Cobo, Francesc Meiras; Dept. Estadística i Investigació Operativa, Universitat de Catalunya.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAppendixReferences |
|---|
[1] Andersen HR, Nielsen JC, Thomsen PEB, Thuesen L, Mortensen PT, Vesterlund T, et al. Long-term follow-up of patients from a randomized trial of atrial versus ventricular pacing for sick-sinus syndrome. Lancet (1997) 350:1210–6.[CrossRef][ISI][Medline]
[2] Gillis AM. Pacing to prevent atrial fibrillation. Cardiol Clin (2000) 18:25–36.[CrossRef][Medline]
[3] Carlson MD, Ip J, Messenger J, Beau S, Kalbfleisch S, Gervais P, et al, Arial dynamic overdrive pacing trial (ADOPT) investigators. A new pacemaker algorithm for the treatment of atrial fibrillation. Results of the atrial dynamic overdrive pacing trial (ADOPT). J Am Coll Cardiol (2003) 42:627–33.
[4] Lewalter T, Yang A, Pfeiffer D, Ruiter J, Schnitzler G, Markert T, et al. Individualized selection of pacing algorithms for the prevention of recurrent atrial fibrillation: results from the VIP registry. Pacing Clin Electrophysiol (2006) 29:124–34.[CrossRef][Medline]
[5] Pürerfellner H, Ruiter JH, Widdershoven JWMNG, Van Gelder IC, Urban L, Kirchhof CJ, et al, PMOP Investigators. Reduction of atrial tachyarrhythmia episodes during the overdrive-pacing period using the post-mode switch overdrive pacing (POMP) algorithm. Heart Rhythm (2006) 31:164–1171.
[6] Daoud EG, Nademanee K, Fuenzalida C, Tomassoni GF, Schuger C, Chisner M, et al. Clinical experience with tiered atrial therapies and atrial arrhythmia prevention algorithms in a dual chamber cardioverter defibrillator. J Cardiovasc Electrophysiol (2006) 17:852–6.[CrossRef][ISI][Medline]
[7] Lee MA, Weachter R, Pollak S, Kremers MS, Naik AM, Silverman R, et al, ATTEST investigators. The effect of atrial pacing therapies on atrial tachyarrhythmia burden and frequency. Results of a randomized trial in patients with bradycardia and atrial tachyarrhythmias. J Am Coll Cardiol (2003) 41:1926–32.
[8] Israel CW, Gronefeld G, Ehrlich JR, Li YG, Hohnloser SH. Prevention of immediate reinitiation of atrial tachyarrhythmias by high-rate overdrive pacing: results from a prospective randomized trial. J Cardiovasc Electrophysiol (2003) 14:954–9.[CrossRef][Medline]
[9] Gregoratos G, Abrans J, Epstein AE, Freedman RA, Hayes DL, HIatky MA, et al. American College of Cardiology/American Heart Association Task Force on Practice Guidelines/North American Society for Pacing and Electrophysiology Committee to Update the 1998 Pacemaker Guidelines. ACC/AHA/NASPE 2002 guideline update for implantation of cardiac pacemakers and antiarrhythmia devices: summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/NASPE Committee to Update the 1998 Pacemaker Guidelines). Circulation (2002) 106:61.[CrossRef]
[10] Israel CW, Hugl B, Unterberg C, Lawo T, Kennis I, Hettrick D, et al, A T500 verification study investigators. Pace-termination and pacing for prevention of atrial tachyarrhythmias: results from a multicenter study with an implantable device for atrial therapy. J Cardiovasc Electrophysiol (2001) 12:1121–8.[CrossRef][ISI][Medline]
[11] Gulizia M, Mangiameli S, Orazi S, Chiaranda G, Boriani G, Piccione G, et al, PITAGORA investigators. Randomized comparison between Ramp and Burst+ atrial antitachycardia pacing therapies in patients suffering from sinus node disease and atrial fibrillation and implanted with a DDDRP device. Europace (2006) 8:465–73.
[12] Israel CW, Ehrlich JR, Gronefeld G, Klesius A, Lawo T, Lemke B, et al. Prevalence, characteristics and clinical implications of regular atrial tachyarrhythmias in patients with atrial fibrillation: insights from a study using a new implantable device. J Am Coll Cardiol (2001) 38:355–63.
[13] Gillis AM, Rose MS. Temporal patterns of paroxysmal atrial fibrillation following DDDR pacemaker implantation. Am J Cardiol (2000) 85:1445–50.[CrossRef][ISI][Medline]
[14] Senn S. Cross-Over Trials in Clinical Research (1993) West Sussex, England: John Wiley & Sons.
[15] Lamas GA, Lee KL, Sweeney MO, Silverman R, Leon A, Yee R, et al. Mode selection trial in sinus node diysfunction. Ventricular pacing or dual chamber pacing for sinus node dysfunction. N Engl J Med (2002) 346:1854–62.
[16] Connolly SJ, Kerr CR, Gent M, Roberts RS, Yusuf S, Gillis AM, et al. Effects of physiologic pacing versus ventricular pacing on the risk of stroke and death due to cardiovascular causes. Canadian trial of physiologic pacing investigators. N Engl J Med (2000) 342:1385–91.
[17] Friedman PA, Dijkman B, Warman EN, Xia HA, Mehra R, Stanton MS, et al. Atrial therapies reduce atrial arrhythmia burden in defibrillator patients. Circulation (2001) 104:1023–8.
[18] Ricci R, Pignalberi C, Disertori M, Capucci A, Padeletti L, Botto G, et al. Efficacy of a dual chamber defibrillator with atrial antitachycardia functions in treating spontaneous atrial tachyarrhythmias in patients with life-threatening ventricular tachyarrhythmias. Eur Heart J (2002) 23:1471–9.
[19] Hemels MEW, Wiesfeld ACP, Inberg B, Van Dessel P, Nieuwland W, Tan ES, et al. Right atrial overdrive pacing for prevention of symptomatic refractory atrial fibrillation. Europace (2006) 8:107–12.
[20] Purerfellner H, Ruiter JH, Widdershoven JW, Van Gelder IC, Urban L, Kirchhof CJ, et al, on behalf of PMOP investigators. Reduction of atrial tachyarrhythmia episodes during the overdrive pacing period using the post-mode switch overdrive pacing (PMOP) algorithm. Heart Rhythm (2006) 3:1164–71.[ISI][Medline]
[21] Padeletti L, Santini M, Boriani G, Botto G, Capucci A, Gulizia M, et al. Temporal variability of atrial tachyarrhythmia burden in bradycardia-tachycardia syndrome patients. Eur Heart J (2005) 26:165–72.
[22] Botto GL, Santini M, Padeletti L, Boriani G, Luzzi G, Zolezzi F, et al. Temporal variability of atrial fibrillation in pacemaker recipients for bradycardia: implications for crossover designed trials, study sample size, and identification of responder patients by means of arrhythmia burden. J Cardiovasc Electrophysiol (2007) 18:250–7.[CrossRef][ISI][Medline]
[23] Chuang-Stein C. The regression fallacy. Drug Inf J (1993) 27:1213–20.
[24] Sweeney M, Bank A, Nsah E, Koullick M, Zeng Q, Hettrick D, et al. Minimizing ventricular pacing to reduce atrial fibrillation in sinus-node disease. N Engl J Med (2007) 357:1000–8.
[25] Lewalter T, Luderitz B. Atrial fibrillation: mortality, morbidity, and money; is that all? Importance and variability of burden measurement. Eur Heart J (2005) 26:110–1.
[26] Blanc J, De Royb L, Mansouratia J, Poezevarac Y, Marcond JL, Schoelse W, et al. Atrial pacing for prevention of atrial fibrillation: assessment of simultaneously implemented algorithms. Europace (2004) 6:371–9.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||