- Email: [email protected]
Association between ventricular pacing and persistent atrial fibrillation in patients indicated to elective pacemaker replacement: Results of the Prefer for Elective Replacement MVP (PreFER MVP) randomized study
Author's Accepted Manuscript
Association between Ventricular Pacing and Persistent Atrial Fibrillation In Patients Indicated to Elective Pacemaker Replacement: Results of the Prefer for Elective Replacement Mvp (Prefer Mvp) Randomized Study Renato P Ricci M.D., Giovanni Luca Botto M.D., Juan M Bénézet M.D., Jens Cosedis Nielsen M.D., Luc De Roy M.D., Olivier Piot M.D., Aurelio Quesada M.D., Raffaele Quaglione M.D., Diego Vaccari M.D., Lorenza Mangoni M.S., Andrea Grammatico Ph.D., Milan Kozák M.D., on behalf of the PreFER MVP Investigators
PII: DOI: Reference:
S1547-5271(15)00815-2 http://dx.doi.org/10.1016/j.hrthm.2015.06.041 HRTHM6340
To appear in:
Heart Rhythm
www.elsevier.com/locate/buildenv
Cite this article as: Renato P Ricci M.D., Giovanni Luca Botto M.D., Juan M Bénézet M. D., Jens Cosedis Nielsen M.D., Luc De Roy M.D., Olivier Piot M.D., Aurelio Quesada M.D., Raffaele Quaglione M.D., Diego Vaccari M.D., Lorenza Mangoni M.S., Andrea Grammatico Ph.D., Milan Kozák M.D., on behalf of the PreFER MVP Investigators, Association between Ventricular Pacing and Persistent Atrial Fibrillation In Patients Indicated to Elective Pacemaker Replacement: Results of the Prefer for Elective Replacement Mvp (Prefer Mvp) Randomized Study, Heart Rhythm, http://dx.doi.org/ 10.1016/j.hrthm.2015.06.041 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Association between ventricular pacing and persistent atrial fibrillation in patients indicated to elective pacemaker replacement: results of the Prefer For Elective Replacement MVP (PreFER MVP) randomized study Renato P Ricci Nielsen
4
1
M.D., Giovanni Luca Botto
M.D., Luc De Roy
Quaglione
8
5
2
M.D., Juan M Bénézet
M.D., Olivier Piot
M.D., Diego Vaccari
9
6
3
M.D., Jens Cosedis
M.D., Aurelio Quesada
M.D., Lorenza Mangoni
10
7
M.D., Raffaele
M.S., Andrea Grammatico
10
Ph.D., Milan Kozák 11 M.D. on behalf of the PreFER MVP Investigators 1 San Filippo Neri Hospital, Rome, Italy; 2 S. Anna Hospital, Como, Italy; 3 Hospital General de Ciudad Real, Ciudad Real, Spain; 4 Aarhus University Hospital, Skejby, Aarhus N, Denmark; 5 Cliniques Universitaires UCL de Mont-Godinne, Yvoir, Belgium; 6 Centre Cardiologique du Nord, Saint-Denis, France; 7 Consorcio Hospital General Universitario de Valencia, Valencia, Spain; 8 Policlinico Umberto I, Rome, Italy; 9 Presidio Ospedaliero di Montebelluna, Montebelluna, Italy; 10 Medtronic EMEA Regional Clinical Centre, Rome, Italy; 11 Fakultni nemocnice Brno Bohunice, Brno, Czech Republic Short title: Ricci - Atrial fibrillation and ventricular pacing after cardiac device replacement Conflict of interest: Renato Ricci received modest consultant honoraria from Biotronik and Medtronic. For other authors, see Disclosure section. Address for correspondence Dr Renato Pietro Ricci, Department of Cardiology San Filippo Neri Hospital, Via Martinotti, 20 00135 Rome, Italy; Phone +390633063934 - email: [email protected]
Abstract Background. Pacing in the right ventricle can cause a variety of detrimental effects, including 1
atrial tachyarrhythmias (AT/AF). Objective. To evaluate incidence and predictors of persistent AT/AF in patients with long term exposure to ventricular pacing. Methods. In a multi-center international trial, 605 patients (75±11 years, 240 females), referred for replacement of an implanted pacemaker or cardioverter defibrillator (ICD), with history of high percentage (>40%) ventricular pacing, were randomly allocated to standard dual-chamber pacing or Managed Ventricular Pacing (MVP), a pacing modality which minimizes ventricular pacing. Main end point of this secondary analysis of the PreFER MVP randomized study was persistent AT/AF, defined as ≥7 consecutive days with AT/AF or AT/AF interrupted by atrial cardioversion or AT/AF present during 2 consecutive follow-up visits. Results. Persistent AT/AF was observed in 71 patients (11.7%) after 2 years of follow-up. At multivariable Cox regression analysis, prior AT/AF (hazard ratio (HR)=2.85, 95% confidence intervals (CI)=1.20-6.22, p=0.017) and ventricular pacing percentage, estimated in the first 3 months, ≥10% (HR=3.24, CI=1.13-9.31, p=0.029) resulted as independent predictors for persistent AT/AF. MVP was associated with persistent AT/AF risk (HR=3.41, CI=1.10-10.6, p=0.024) in the subgroup of patients with a baseline long PR interval (PR>230 ms), but not in the whole population. Conclusion. In pacemaker and ICD replacement patients, high percentage of ventricular pacing is associated with higher risk of persistent AT/AF. The use of algorithms which minimize right ventricular pacing may benefit patients with normal spontaneous AV conduction but should be evaluated with caution in patients with long PR interval. Keywords: Cardiac pacing; Managed Ventricular Pacing; Randomized controlled trial; Atrial fibrillation; PR interval.
Clinical Trial Registration: URL http://www.clinicaltrials.gov. Identifier: NCT00293241 Abbreviations AF=atrial fibrillation 2
AAI=single-chamber atrial pacing mode AT/AF=atrial tachyarrhythmia CI= 95% confidence intervals DDD=dual-chamber pacing mode HF=heart failure HR= hazard ratios ICD=implantable cardioverter defibrillator IQR=interquartile range LVEF=left ventricle ejection fraction MVP=managed ventricular pacing NYHA=New York Heart Association SD=standard deviation Introduction Atrial fibrillation (AF) is a frequent condition in patients with implantable cardiac devices, being present in about one fourth of patients at first device implant and in up to half of patients during follow-up (1-2). AF has been associated with increased mortality and morbidity and with impairment of quality of life (3). The best pacing modality to reduce AF incidence is still being debated (4-10). A number of clinical studies (4-9) have shown that, in patients with intact AV conduction, unnecessary chronic right-ventricular pacing can cause a variety of detrimental effects, including AF and congestive heart failure (HF). These effects are believed to result from the mechanical
3
dyssynchrony and ventricular chamber dysfunction that occurs with chronic, single-site, apical ventricular stimulation. Managed Ventricular Pacing (MVP), has been designed to give preference to intrinsic ventricular activation by minimizing ventricular pacing (11-12). This is accomplished by an atrial-based dual-chamber pacing mode that provides functional single-chamber atrial (AAI) pacing with ventricular monitoring and automatic switch from AAI to dual-chamber (DDD) pacing during episodes of AV block. The PreFER MVP study (13) was designed to compare standard DDD pacing with MVP in patients with pacemaker or implantable cardioverter defibrillator (ICD) who needed device replacement. This population is of particular interest because of its long pacing history and because it represents an important portion of the total implants, ranging from 13% to 21%. (14) The results of PreFER MVP study in terms of primary end point, showed no significant difference in cardiovascular hospitalizations between the two study groups (15). Aim of this predefined secondary analysis of PreFER MVP study was to evaluate the incidence of persistent atrial tachyarrhythmias (AT/AF) in relation to patient characteristics and pacing modality in patients indicated for device replacement. Methods Study design and patient population The design of the PreFER MVP study has been previously described (13, 15). In brief, the study was a global, prospective, randomized, single-blinded multi-center trial. A total of 630 patients, planned to have their dual-chamber pacemaker or ICD replaced, were enrolled in 76 centers in Europe, Canada, Australia, Israel, Kuwait, Hong Kong and South Korea. Enrollment took place between February 21, 2006, and August 19, 2009.
4
The trial was conducted in accordance with the Helsinki Declaration. The Institution Review Board or Medical Ethic Committee approved the study protocol and all patients provided their informed consent. Inclusion criteria required that patients had been implanted with a dual-chamber device for a minimum duration of 2 years, received more than 40% ventricular pacing as detected by the device over a period of >4 weeks before replacement and that there was no clinical need to change the pacing mode or the AV intervals. Exclusion criteria are detailed elsewhere (see ref 13) and included cardiac resynchronization therapy indication, permanent AF, permanent 3rd degree AV Block, age less than 18 years, and neurocardiogenic syncope as primary pacemaker indication. Randomization was provided by an electronic database and was stratified by left ventricular ejection fraction (LVEF) - ≤40% or >40% - and by pacemaker versus ICD.
Follow-up and study end points Clinical information was retrieved at baseline and at scheduled follow-up visits at 1 month, 12 months and 24 months. After the 2 years follow-up visit, patients remained in the study and clinical and device data were collected until the last enrolled subject reached 24 months of follow-up. Daily total AT/AF duration as detected by the device was retrieved from device memory. The main objective of this secondary analysis was to identify predictors of persistent AT/AF among baseline patient characteristics and pacing modalities. The primary end point was persistent AT/AF, defined as either longer than 7 consecutive days with device diagnostics
5
showing 20 or more hours in AT/AF or AT/AF interrupted by atrial cardioversion or AT/AF present at 2 consecutive follow-up visits. This analysis was pre-specified in the study protocol and in the statistical analysis plan. We also measured other AT/AF-related end points such as the first occurrence of at least 5 minutes, 1 hour and 6 hours of AT/AF, and permanent AT/AF defined as one of the following: 7 consecutive days with device diagnostics showing AT/AF for ≥20 hours and either failed cardioversion or the investigator decided not to cardiovert the patient. Device Programming Once a patient was randomized, the following programming requirements were defined: patients in the DDD arm should be programmed as in the exchanged device, without a change in sensed and paced AV intervals, and no features switched on which had not been available or switched on in the previous device, while patients in the MVP arm should be programmed in AAI(R)<=>DDD(R) pacing mode, with Sensed AV interval, Paced AV interval and Rate adaptive AV left to physician discretion. Mode switch and AT/AF detection were required to be enabled in both study arms.
Medical Treatment Medical treatment was left to physician discretion and investigators were requested to treat HF patients with the optimal medical therapy available as described in the ESC Guidelines for the diagnosis and treatment of chronic HF (16). Statistical analysis Descriptive statistics were reported as mean and standard deviation for normally distributed continuous variables, or median with 25th-75th percentile interquartile range (IQR) in the case of skewed distribution. The incidence of end points at 2 years follow-up was estimated by the
6
Kaplan–Meier method and compared with the log-rank test. Cox regression was used to calculate hazard ratio (HR) with 95% confidence interval (CI). Univariable and multivariable Cox regression analysis was used to identify risk factors for persistent AT/AF occurrence. Two Cox regression models were estimated, one using the median PR interval and one using the upper quartile PR interval value. Beside baseline patient’s characteristics, also ventricular pacing percentage, as retrieved from device memory, was tested as a predictor of persistent AT/AF. We chose to use ventricular pacing percentage as estimated in the first 3 months of each patient observation period, in order to characterize the patient with a variable measured in a follow-up period as near as possible to baseline evaluation. Patients were deemed to have high ventricular pacing percentage if their ventricular pacing percentage was equal or higher than 10% and were compared with patients with ventricular pacing percentage lower than 10%. The main analysis considered all occurrences of persistent AT/AF. A sensitivity analysis was also performed excluding persistent AT/AF events which occurred in the first 3 months, to account for the fact that baseline ventricular pacing was estimated in the first 3 follow-up months after implant. The 2 year incidence of persistent AT/AF was estimated also in patients sub-groups selected as a function of PR interval, ventricular pacing percentage and randomization arm. Statistical tests were two-tailed, and p<0.05 was considered significant. The analysis adhered to the intent-to-treat principle. All the statistical analyses were performed using SAS 9.3. Results All 605 patients randomized in the PreFER MVP study were included in the analysis. Baseline patient characteristics are described in table I. In total, 186 (30.7%) patients had a known history of AF and 43 (7.1%) patients had a history of atrial tachycardia or atrial flutter.
7
Mean follow-up was 2.2 ± 1.0 years. Clinically defined persistent AT/AF Clinically defined persistent AT/AF was observed in 71 patients (actuarial incidence: 13.2%, CI=10.6%-16.5%) at 2 years. In univariable Cox regression analysis, persistent AT/AF was significantly associated with prior AF, coronary artery disease, cardiomyopathy, use of antiarrhythmic drugs, LVEF, ventricular pacing percentage and PR interval, but not with the randomization group, as shown in table II. In particular, the risk of developing persistent AT/AF for the whole study population increased by a factor 4.09 (HR=4.09, CI=2.11-7.94, p<0.001) in patients with ventricular pacing percentage higher than 10%. These patients represent two third of the whole patient population as shown in the ventricular pacing percentage distribution (figure 1). Median ventricular pacing percentage was 53.4% (lower-upper quartile 2.7%-96.0%) in the whole population, 88.0% (lower-upper quartile 44.5%-98.4%) in the DDD arm and 5.1% (lowerupper quartile 0.5%-64.2%) in the MVP arm. The association between persistent AT/AF and baseline PR interval was found for PR intervals longer or equal to 230 ms, corresponding to the upper quartile PR value in the studied population (HR =1.84, CI=1.00-3.37, p=0.049) while it was not significant (HR =1.35, CI=0.71-2.56, p=0.355) for PR≥210 ms, corresponding to the median PR interval in the studied population. Figure 2 shows time to first persistent AT/AF for the entire population in patients with ventricular pacing percentage ≤10% (continuous line) and >10% (dotted line). Figure 3 shows time to first persistent AT/AF in patients with PR interval < 230 ms (continuous line) and ≥ 230 ms (dotted line). Multivariable Cox regression analysis (table II) demonstrated that only a history of prior AT/AF (HR =2.85, CI=1.20-6.22, p=0.017) and high ventricular pacing percentage (HR =3.24, CI=1.13-
8
9.31, p=0.029) were independent predictors for persistent AT/AF. No qualitative differences were identified by repeating Cox Regression model on the sensitivity analyses. Secondary end points In the first 2 years post randomization 44.3% of the patients experienced 1 or more days with at least 5 minutes of device-detected AT/AF, 35.7% with at least 1 hour of AT/AF, 18.8% with at least 6 hours of AT/AF, while 7.9% of the patients had AT/AF longer than 7 days and 3.5% of patients were considered having permanent AT/AF. The randomization arm was not associated with statistically significant differences for any of the described AT/AF conditions when evaluating the whole study population. In the subgroup of patients with long PR interval (PR≥230 ms) MVP was related to a higher incidence of persistent AT/AF when compared with DDD pacing (HR 3.41, CI=1.10-10.6, p=0.024). In the subgroup of patients with shorter PR interval (PR<230 ms), patients randomized to MVP mode and having a ventricular pacing percentage lower than the median pacing percentage were associated with lower incidence of persistent AT/AF either when compared with patients randomized to DDD pacing and having a ventricular pacing percentage lower than the median pacing percentage (HR 0.25, CI=0.067-0.900, p=0.034) or when compared with patients randomized to MVP pacing and having a ventricular pacing percentage higher than the median pacing percentage. In patients randomized to the MVP arm, the median (IQR) ventricular pacing percentage was 50.5 % (10.9%-97.2%) in patients who developed persistent AT/AF and 2.7% (0.4%47.8%) in patients who did not develop persistent AT/AF.Discussion Main results The main result of this secondary analysis of PreFER MVP study was that in patients referred for replacement of an implanted pacemaker or ICD, prior AT/AF and high ventricular pacing 9
percentage were independent predictors of persistent AT/AF during the follow-up. In patients with long PR intervals (PR≥230 ms) MVP mode was associated with higher risk of persistent AT/AF compared with DDD mode. In patients with shorter PR intervals (PR<230 ms) MVP mode was associated with lower risk of persistent AT/AF compared with DDD mode, in patients with a low percentage of ventricular pacing. Atrial fibrillation prevalence in patients wearing implantable cardiac devices Our data confirm previous findings about the high prevalence of AT/AF in pacemaker and defibrillator patients [1-3]. A history of AT/AF was present in 37.8% of patients at baseline and these arrhythmias were documented in the device memories in 44.3% of patients during the 2year observation period; in real clinical practice AT/AF prevalence has to be estimated even higher when including permanent AT/AF which was an exclusion criterion in our study. Predictors of persistent atrial fibrillation In our patient population the development of persistent AT/AF was significantly associated with high ventricular pacing percentage (table II and figure 2). This finding confirms results of previous studies (4-7, 9). Some small clinical studies (4-6) and the post-hoc analyses of the MOST trial (7) suggested the association between the risk to develop AT/AF and high ventricular pacing percentage. The SAVEPACe trial (9), in 1065 patients with sinus-node disease, intact atrioventricular conduction, and normal QRS interval, compared conventional dual-chamber pacing with dual-chamber minimal ventricular pacing using pacemaker features designed to promote spontaneous atrioventricular conduction and to minimize ventricular pacing. The results showed that the incidence of persistent AT/AF was significantly reduced by 40% (p=0.009) in the group of patients assigned to minimal ventricular pacing compared with the DDD arm, in which the mean right ventricular pacing was 99%.
10
In our study, persistent AT/AF was not reduced in the MVP arm of the PreFER MVP trial, despite the fact that the MVP mode significantly reduced ventricular pacing percentage compared with DDD mode. This represents an intriguing issue, for which we propose the following possible explanations. First, PreFER MVP (15), by study design, selected a population of patients who were paced more than 40% of time by previous pacemaker or ICD and who survived their first implant without developing permanent AF and were therefore more predisposed to tolerate long term ventricular pacing. Secondly, risk for persistent AT/AF strongly increased for ventricular pacing percentage higher than 10%, and in the MVP arm of the study, in spite of a median value of 5.1%, a proportion of patients continued to be highly paced in the ventricle (upper quartile was 64.2%). Interestingly, in patients randomized to the MVP arm, the median ventricular pacing percentage was 50.5 % in patients who developed persistent AT/AF and 2.7% in patients who did not develop persistent AT/AF. Finally, persistent AT/AF was significantly associated with long PR (PR≥230 ms) and this finding may have overwhelmed the benefits of ventricular pacing reduction in MVP patients. As a matter of fact, in the DANPACE trial (10, 17), which randomly assigned 1415 sick sinus syndrome patients referred for first pacemaker implantation to AAIR (707) or DDDR (708) pacing, followed for a mean of 5 years, an excess of AF was observed in patients paced with single lead atrial pacing as compared with those paced in DDD mode, in particular in patients with prolonged intrinsic PR interval. Clinical implications As reported by Botto et al. (15), the PreFER MVP study aimed to compare MVP and DDD pacing modes in patients indicated to elective pacemaker or ICD replacement with high prior ventricular pacing percentage. The main analysis showed that in these patients, with clinically well tolerated long term exposure to more than 40% pacing in the ventricle, a strategy to
11
minimize ventricular pacing is not superior in reducing incidence of cardiovascular hospitalization compared to standard DDD pacing. Findings of the present analysis showing the association between ventricular pacing percentage greater than 10% and persistent AT/AF confirms the favorable role of pacing modalities to reduce unnecessary ventricular pacing in selected patients with normal spontaneous AV conduction. Reducing ventricular pacing below 10% may require dedicated algorithms and may be difficult to obtain by simply programming long AV interval. Indeed, subgroup analyses in our study showed that patients with long baseline PR interval (PR≥230 ms) had a higher incidence of persistent AT/AF if programmed in MVP compared with DDD pacing. We hypothesize that enabling very long intrinsic atrioventricular conduction times, as in the MVP arm, may cause AT/AF in patients with long PR intervals who were used to being chronically paced in the ventricle for a long time. As reported earlier (17-18), a long PR interval could be responsible for a higher incidence of AF, either in patients with or without history of AF. The actual reason of this association remains still hypothetical. Conversely, patients with shorter PR interval (PR<230 ms) randomized to MVP mode and having a ventricular pacing percentage lower than 5.1% (median pacing percentage in MVP mode) showed lower persistent AT/AF either compared with patients in DDD mode and low ventricular pacing percentage or patients in MVP mode and high ventricular pacing percentage. These data suggest that MVP is an effective pacing mode in patients with normal or only slightly prolonged PR interval, while pacing at a relatively short AV interval may be better in patients with a prolonged PR interval. When comparing MVP and DDD pacing, clinical benefit may derive from the balance of two conflicting factors, AV synchrony and ventricular pacing percentage. Although DDD pacing guarantees AV synchrony, unnecessary pacing in the right ventricle may cause ventricular remodeling due to the consequent change in electrical activation and contraction pattern of the
12
ventricles. Conversely, MVP preserves a normal ventricular contraction pattern, which is expected to reduce ventricular remodeling in comparison with a higher percentage of ventricular pacing, but at the cost of allowing a prolonged AV conduction in some patients. In patients undergoing pacemaker or ICD replacement with high percentage ventricular pacing history, the choice of the best pacing mode should be guided by careful evaluation of several clinical factors, including ventricular pacing percentage in the previous device and spontaneous PR interval. Our data suggest that programming the MVP mode is of major importance in reducing the percentage of ventricular pacing and hence the incidence of AF. Only in patients with a prolonged PR interval (≥ 230 ms) a DDD mode, with pacing at a relatively short AV interval, should be recommended. Study limitations We report results of secondary analyses of the PreFER MVP study. The statistical methods of the analyses on the main endpoint, persistent AT/AF, were pre-specified in the study protocol and in the statistical analysis plan, with the only exemption of the ventricular pacing percentage cut-off used to evaluate the association of this variable with persistent AT/AF. In particular, the cut-off (10%) used to divide the patient population between patients with low vs. high ventricular pacing percentage was chosen since it was the distribution tertile and possibly a good candidate to differentiate pacing modes devoted to minimize ventricular pacing from standard DDD pacing based on the observation of the skewed distribution of ventricular pacing percentage as measured in our study (figure 1) and in previous trials. (11-12, 19-20) The primary end point, persistent AT/AF, consists of three components, new-onset persistent AT/AF in patients without previous history of AT/AF, persistent AT/AF in patients with history of paroxysmal AT/AF, and recurrence of persistent AT/AF in patients with history of persistent
13
AT/AF. We cannot exclude that different mechanisms are responsible for the development of persistent AT/AF in these different situations. The use of antiarrhythmic drugs could have interfered with arrhythmia occurrence. The results of our analyses should be regarded as characteristics of the studied population of pacemaker and ICD replacement patients, with clinically well tolerated long term exposure to more than 40% pacing in the ventricle. In particular the results refer to conventional right atrial appendage pacing. We can not exclude that different results might have been observed in case of alternative site atrial pacing. Conclusions In pacemaker and ICD replacement patients, high percentage of ventricular pacing is associated with higher risk of persistent AT/AF. The use of algorithms which minimize unnecessary right ventricular pacing may benefit patients with normal spontaneous AV conduction but should be evaluated with caution in patients with long PR interval. Funding The PreFER MVP study was supported and sponsored by Medtronic Bakken Research Center. Acknowledgements The authors thank Bart Gerritse and Lidwien Vainer for contributing to and reviewing scientific and statistical methods and manuscript draft. Disclosures Renato Ricci has received consultant honoraries from Biotronik and Medtronic, and speaker fees from Biotronik. Giovanni Luca Botto has received research grants from Boston Scientific, St. Jude Medical, Bayer Healthcare, Gilead, Sanofi, Medtronic, consultant honoraries from Biotronik, Boston Scientific, St. Jude Medical, MSD, Bayer Healthcare, Sanofi, Medtronic, and speaker fees from Boston 14
Scientific, St. Jude Medical, Bayer Healthcare, Boheringer, Sanofi, Sorin, Pfizer, MSD and Medtronic. Juan M Bénézet has received consultant honoraries and speaker fees from Medtronic. Jens Cosedis Nielsen has received speaker fees from Biotronik and Biosense Webster, consultant honoraries from Boston Scientific, and research grants for the MANTRA-PAF trial from Biosense Webster. Luc De Roy received research and/or fellowship funds or consultant fees from Medtronic, St Jude, Boston, Sorin and Biotronic. Olivier Piot has received consultant honoraries from St Jude Medical, Sorin, Biotronik and Medtronic.
Aurelio Quesada has received research grants from Sorin
and Medtronic. Diego Vaccari has received research grants from St. Jude Medical, and speaker fees from Biotronik. Milan Kozak received consultant honoraries from Boston Scientific, Medtronic and Biotronik. Lorenza Mangoni and Andrea Grammatico are employee of the Medtronic EMEA Regional Clinical Centre. References 1. Israel CW, Grönefeld G, Ehrlich JR, Li YG, Hohnloser SH. Long-term risk of recurrent atrial fibrillation as documented by an implantable monitoring device: implications for optimal patient care. J Am Coll Cardiol. 2004;43:47-52. 2. Healey JS, Connolly SJ, Gold MR et al. ASSERT Investigators. Subclinical atrial fibrillation and the risk of stroke. N Engl J Med. 2012;366:120-9. 3. Fuster V, Rydén LE, Cannom DS et al. American College of Cardiology Foundation/American Heart Association Task Force. 2011 ACCF/AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. Circulation. 2011;123:e269-367.
15
4. Andersen HR, Cosedis Nielsen J, Thomsen PE, Thuesen L, Mortensen PT, Vesterlund T, Pedersen AK. Long-term follow-up of patients from a randomised trial of atrial versus ventricular pacing for sick-sinus syndrome. Lancet 1997; 350:1210–6. 5. Cosedis Nielsen J, Kristensen L, Andersen HR, Mortensen PT, Pedersen OL, Pedersen AK. A randomized comparison of atrial and dual chamber pacing in 177 consecutive patients with sick sinus syndrome: echocardiographic and clinical outcome. J Am Coll Cardiol 2003; 42:614–23. 6. Kristensen L, Cosedis Nielsen J, Mortensen PT, Pedersen OL, Pedersen AK, Andersen HR. Incidence of atrial fibrillation and thromboembolism in a randomised trial of atrial versus dual chamber pacing in 177 patients with sick sinus syndrome. Heart 2004; 90:661–6. 7. Sweeney MO, Hellkamp AS, Ellenbogen KA, Greenspon AJ, Freedman RA, Lee KL, Lamas GA; MOde Selection Trial Investigators. 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; 107:2932–7. 8. Gillis AM. Redefining physiologic pacing: lessons learned from recent clinical trials. Heart Rhythm 2006; 3:1367-72. 9. Sweeney MO, Bank AJ, Nsah E, Koullick M, Zeng QC, Hettrick D, Sheldon T, Lamas GA; Search AV Extension and Managed Ventricular Pacing for Promoting Atrioventricular Conduction (SAVE PACe) Trial. Minimizing Ventricular Pacing to Reduce Atrial Fibrillation in Sinus-Node Disease. N Engl J Med 2007; 357:1000-8.
16
10. Cosedis Nielsen J, Thomsen PE, Højberg S et al. DANPACE Investigators. A comparison of single-lead atrial pacing with dual-chamber pacing in sick sinus syndrome. Eur Heart J 2011; 32:686–96. 11. Sweeney MO, Shea JB, Fox V, Adler S, Nelson L, Mullen TJ, Belk P, Casavant D, Sheldon T. Randomized pilot study of a new atrial-based minimal ventricular pacing mode in dual-chamber implantable cardioverter-defibrillators. Heart Rhythm 2004; 1:160-7 12. Gillis AM, Pürerfellner H, Israel CW, Sunthorn H, Kacet S, Anelli-Monti M, Tang F, Young M, Boriani G; Medtronic Enrhythm Clinical Study Investigators. Reducing unnecessary right ventricular pacing with the managed ventricular pacing mode in patients with sinus node disease and AV block. Pacing Clin Electrophysiol 2006; 29:697705. 13. Quesada A, Botto G, Erdogan A, Kozak M, Lercher P, Cosedis Nielsen J, Piot O, Ricci R, Weiss C, Becker D, Wetzels G, De Roy L; PreFER MVP Investigators. Managed ventricular pacing vs. conventional dual-chamber pacing for elective replacements: the PreFER MVP study: clinical background, rationale, and design. Europace 2008; 10:321– 6. 14. Kurtz SM, Ochoa JA, Lau E, Shkolnikov Y, Pavri BB, Frisch D, Greenspon AJ. Implantation trends and patient profiles for pacemakers and implantable cardioverter defibrillators in the United States: 1993-2006. Pacing Clin Electrophysiol 2010; 33:70511. 15. Botto GL, Ricci RP, Bénézet JM, Cosedis Nielsen J, De Roy L, Piot O, Quesada A, Quaglione R, Vaccari D, Garutti C, Vainer L, Kozák M, on behalf of the PreFER MVP Investigators, Managed Ventricular Pacing compared to conventional dual-chamber pacing for elective replacement in chronically paced patients: results of the Prefer for 17
Elective Replacement MVP (PreFER MVP) randomized study. Heart Rhythm 2014; 11:992-1000. 16. Swedberg K, Cleland J, Dargie H et al. Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (updated 2005). Eur Heart J 2005; 26:1115-40. 17. Cosedis Nielsen J, Thomsen PE, Højberg S et al. DANPACE investigators. Atrial fibrillation in patients with sick sinus syndrome: the association with PQ-interval and percentage of ventricular pacing. Europace 2012; 14:682-9. 18. Cheng S, Keyes MJ, Larson MG, McCabe EL, Newton-Cheh C, Levy D, Benjamin EJ, Vasan RS, Wang TJ. Long-term outcomes in individuals with prolonged PR interval or first-degree atrioventricular block. JAMA 2009; 301:2571-7. 19. Almendral J, Arribas F, Wolpert C, Ricci R, Adragao P, Cobo E, Navarro X, Quesada A; DATAS Steering Committee; DATAS Writing Committee; DATAS Investigators. Dualchamber defibrillators reduce clinically significant adverse events compared with singlechamber devices: results from the DATAS (Dual chamber and Atrial Tachyarrhythmias Adverse events Study) trial. Europace. 2008;10(5):528-35 20. Barsheshet A, Moss AJ, McNitt S, Jons C, Glikson M, Klein HU, Huang DT, Steinberg JS, Brown MW, Zareba W, Goldenberg I; MADIT-II Executive Committee. Long-term implications of cumulative right ventricular pacing among patients with an implantable cardioverter-defibrillator. Heart Rhythm. 2011;8(2):212-8. Clinical perspectives This secondary analysis of the PreFER MVP clinical trial provides new evidence about the fact that in patients undergoing pacemaker or ICD replacement, high percentage of
18
ventricular pacing is associated with higher risk of developing persistent AT/AF. Our data suggest that the use of algorithms minimizing unnecessary right ventricular pacing is warranted in patients who have normal spontaneous AV conduction while it should be evaluated with caution in patients with long PR interval. These finding have clinical importance because patients undergoing pacemaker or ICD replacement who have a high percentage ventricular pacing history, represent a relevant portion of total implants. So far no data have been published about the choice of the best pacing mode in this specific patientpopulation.
19
Table I: Baseline patients characteristics
Subject Characteristics Male gender N (%) Age (years) Implanted device - Pacemaker (N,%) Implanted device - ICD (N,%)
Total (N = 605) 365 (60.3%) 75 ± 11 556 (91.9%) 49 (8.1%)
Primary Device (pacemaker only) Indication Sinus Node Dysfunction N (%)
372 (61.5%)
AV block N (%)
139 (23.0%)
Other brady indication N (%) Previous device implant duration (years) - mean ± SD Ventricular pacing percentage in the period preceding study enrollment - median (IQR)
45 (7.4%) 7.7 + 3.3 95% (IQR 75% – 99%)
History of Atrial fibrillation N (%)
186 (30.7%)
Paroxysmal N (%)
168 (27.8%)
Persistent N (%)
18 (3.1%)
History of Atrial flutter/tachycardia N (%)
43 (7.1%)
Cardiomyopathy N (%)
118 (19.5%)
Coronary Artery Disease N (%)
140 (23.1%)
Hypertension N (%)
307 (50.7%)
Diabetes mellitus N (%)
101 (16.7%)
Heart Failure N (%)
434 (71.7%)
NYHA Class I N (%)
186 (30.7%)
NYHA Class II N (%)
216 (35.7%)
NYHA Class III-IV N (%) LVEF (%) – mean ± SD Intrinsic P-R Interval (ms) median (IQR)
32 (5.3%) 53 ± 13 210 (170-230)
Cardiovascular Medications N (%) Aspirin/Acetylsalicylic acid N (%)
296 (48.9%)
Anticoagulants N (%)
147 (24.3%)
Diuretics N (%)
196 (32.4%)
Angiotensin-converting enzyme inhibitors N (%)
214 (35.4%)
Angiotensin II receptor blockers N (%)
112 (18.5%)
Beta-blockers N (%)
245 (40.5%)
Calcium antagonists N (%)
117 (19.3%)
Digitalis/digoxin N (%) Amiodarone N (%)
56 (9.3%) 100 (16.5%)
LVEF=left ventricle ejection fraction, NYHA=New York Heart Association, SD=standard deviation, IQR=25th-75th interquartile range
20
Table II Cox regression model for predictors of persistent AT/AF
Univariate
Multivariate
p value
Hazard Ratio
95% Hazard Ratio Confidence Limits
p value
Baseline Characteristic
Hazard Ratio
95% Hazard Ratio Confidence Limits
Prior AT/AF
3.52
2.30-5.38
<0.001
2.85
1.20-6.22
0.017
Coronary artery disease
1.95
1.26-3.01
0.003
1.19
0.21-6.76
0.848
Cardiomyopathy
2.20
1.41-3.41
<0.001
1.43
0.46-4.40
0.533
Antiarrhythmic drugs
2.50
1.59-3.94
<0.001
1.66
0.68-4.09
0.267
LVEF (%)
0.97
0.97-0.99
0.001
0.99
0.95-1.03
0.557
PR interval (≥230 ms)
1.84
1.00-3.37
0.049
2.11
0.87-5.10
0.097
1.35
0.88-2.05
0.167
2.43
0.95-6.22
0.063
4.09
2.11-7.94
<0.001
3.24
1.13-9.31
0.029
Randomization [MVP ON / MVP OFF] Ventricular pacing percentage (estimated in the first 3 months – cut off ≥10%)
Figure legend Figure 1 Distribution of ventricular pacing percentage as measured in the first 3 months of the observation period in the whole population
21
Figure 2 Time to first persistent AT/AF in patients with ventricular pacing percentage ≤ 10% (continuous line) and > 10% (dotted line) in the whole population, for the subgroup of 549 patients with complete data about ventricular pacing percentage. Figure 3 Time to first persistent AT/AF in patients with PR interval < 230 ms (continuous line) and ≥ 230 ms (dotted line) in the whole population, for the subgroup of 353 patients with complete data about PR interval.
22
Figure 1
Figure 2
Figure 3
Association between Ventricular Pacing and Persistent Atrial Fibrillation In Patients Indicated to Elective Pacemaker Replacement: Results of the Prefer for Elective Replacement Mvp (Prefer Mvp) Randomized Study Renato P Ricci M.D., Giovanni Luca Botto M.D., Juan M Bénézet M.D., Jens Cosedis Nielsen M.D., Luc De Roy M.D., Olivier Piot M.D., Aurelio Quesada M.D., Raffaele Quaglione M.D., Diego Vaccari M.D., Lorenza Mangoni M.S., Andrea Grammatico Ph.D., Milan Kozák M.D., on behalf of the PreFER MVP Investigators
PII: DOI: Reference:
S1547-5271(15)00815-2 http://dx.doi.org/10.1016/j.hrthm.2015.06.041 HRTHM6340
To appear in:
Heart Rhythm
www.elsevier.com/locate/buildenv
Cite this article as: Renato P Ricci M.D., Giovanni Luca Botto M.D., Juan M Bénézet M. D., Jens Cosedis Nielsen M.D., Luc De Roy M.D., Olivier Piot M.D., Aurelio Quesada M.D., Raffaele Quaglione M.D., Diego Vaccari M.D., Lorenza Mangoni M.S., Andrea Grammatico Ph.D., Milan Kozák M.D., on behalf of the PreFER MVP Investigators, Association between Ventricular Pacing and Persistent Atrial Fibrillation In Patients Indicated to Elective Pacemaker Replacement: Results of the Prefer for Elective Replacement Mvp (Prefer Mvp) Randomized Study, Heart Rhythm, http://dx.doi.org/ 10.1016/j.hrthm.2015.06.041 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Association between ventricular pacing and persistent atrial fibrillation in patients indicated to elective pacemaker replacement: results of the Prefer For Elective Replacement MVP (PreFER MVP) randomized study Renato P Ricci Nielsen
4
1
M.D., Giovanni Luca Botto
M.D., Luc De Roy
Quaglione
8
5
2
M.D., Juan M Bénézet
M.D., Olivier Piot
M.D., Diego Vaccari
9
6
3
M.D., Jens Cosedis
M.D., Aurelio Quesada
M.D., Lorenza Mangoni
10
7
M.D., Raffaele
M.S., Andrea Grammatico
10
Ph.D., Milan Kozák 11 M.D. on behalf of the PreFER MVP Investigators 1 San Filippo Neri Hospital, Rome, Italy; 2 S. Anna Hospital, Como, Italy; 3 Hospital General de Ciudad Real, Ciudad Real, Spain; 4 Aarhus University Hospital, Skejby, Aarhus N, Denmark; 5 Cliniques Universitaires UCL de Mont-Godinne, Yvoir, Belgium; 6 Centre Cardiologique du Nord, Saint-Denis, France; 7 Consorcio Hospital General Universitario de Valencia, Valencia, Spain; 8 Policlinico Umberto I, Rome, Italy; 9 Presidio Ospedaliero di Montebelluna, Montebelluna, Italy; 10 Medtronic EMEA Regional Clinical Centre, Rome, Italy; 11 Fakultni nemocnice Brno Bohunice, Brno, Czech Republic Short title: Ricci - Atrial fibrillation and ventricular pacing after cardiac device replacement Conflict of interest: Renato Ricci received modest consultant honoraria from Biotronik and Medtronic. For other authors, see Disclosure section. Address for correspondence Dr Renato Pietro Ricci, Department of Cardiology San Filippo Neri Hospital, Via Martinotti, 20 00135 Rome, Italy; Phone +390633063934 - email: [email protected]
Abstract Background. Pacing in the right ventricle can cause a variety of detrimental effects, including 1
atrial tachyarrhythmias (AT/AF). Objective. To evaluate incidence and predictors of persistent AT/AF in patients with long term exposure to ventricular pacing. Methods. In a multi-center international trial, 605 patients (75±11 years, 240 females), referred for replacement of an implanted pacemaker or cardioverter defibrillator (ICD), with history of high percentage (>40%) ventricular pacing, were randomly allocated to standard dual-chamber pacing or Managed Ventricular Pacing (MVP), a pacing modality which minimizes ventricular pacing. Main end point of this secondary analysis of the PreFER MVP randomized study was persistent AT/AF, defined as ≥7 consecutive days with AT/AF or AT/AF interrupted by atrial cardioversion or AT/AF present during 2 consecutive follow-up visits. Results. Persistent AT/AF was observed in 71 patients (11.7%) after 2 years of follow-up. At multivariable Cox regression analysis, prior AT/AF (hazard ratio (HR)=2.85, 95% confidence intervals (CI)=1.20-6.22, p=0.017) and ventricular pacing percentage, estimated in the first 3 months, ≥10% (HR=3.24, CI=1.13-9.31, p=0.029) resulted as independent predictors for persistent AT/AF. MVP was associated with persistent AT/AF risk (HR=3.41, CI=1.10-10.6, p=0.024) in the subgroup of patients with a baseline long PR interval (PR>230 ms), but not in the whole population. Conclusion. In pacemaker and ICD replacement patients, high percentage of ventricular pacing is associated with higher risk of persistent AT/AF. The use of algorithms which minimize right ventricular pacing may benefit patients with normal spontaneous AV conduction but should be evaluated with caution in patients with long PR interval. Keywords: Cardiac pacing; Managed Ventricular Pacing; Randomized controlled trial; Atrial fibrillation; PR interval.
Clinical Trial Registration: URL http://www.clinicaltrials.gov. Identifier: NCT00293241 Abbreviations AF=atrial fibrillation 2
AAI=single-chamber atrial pacing mode AT/AF=atrial tachyarrhythmia CI= 95% confidence intervals DDD=dual-chamber pacing mode HF=heart failure HR= hazard ratios ICD=implantable cardioverter defibrillator IQR=interquartile range LVEF=left ventricle ejection fraction MVP=managed ventricular pacing NYHA=New York Heart Association SD=standard deviation Introduction Atrial fibrillation (AF) is a frequent condition in patients with implantable cardiac devices, being present in about one fourth of patients at first device implant and in up to half of patients during follow-up (1-2). AF has been associated with increased mortality and morbidity and with impairment of quality of life (3). The best pacing modality to reduce AF incidence is still being debated (4-10). A number of clinical studies (4-9) have shown that, in patients with intact AV conduction, unnecessary chronic right-ventricular pacing can cause a variety of detrimental effects, including AF and congestive heart failure (HF). These effects are believed to result from the mechanical
3
dyssynchrony and ventricular chamber dysfunction that occurs with chronic, single-site, apical ventricular stimulation. Managed Ventricular Pacing (MVP), has been designed to give preference to intrinsic ventricular activation by minimizing ventricular pacing (11-12). This is accomplished by an atrial-based dual-chamber pacing mode that provides functional single-chamber atrial (AAI) pacing with ventricular monitoring and automatic switch from AAI to dual-chamber (DDD) pacing during episodes of AV block. The PreFER MVP study (13) was designed to compare standard DDD pacing with MVP in patients with pacemaker or implantable cardioverter defibrillator (ICD) who needed device replacement. This population is of particular interest because of its long pacing history and because it represents an important portion of the total implants, ranging from 13% to 21%. (14) The results of PreFER MVP study in terms of primary end point, showed no significant difference in cardiovascular hospitalizations between the two study groups (15). Aim of this predefined secondary analysis of PreFER MVP study was to evaluate the incidence of persistent atrial tachyarrhythmias (AT/AF) in relation to patient characteristics and pacing modality in patients indicated for device replacement. Methods Study design and patient population The design of the PreFER MVP study has been previously described (13, 15). In brief, the study was a global, prospective, randomized, single-blinded multi-center trial. A total of 630 patients, planned to have their dual-chamber pacemaker or ICD replaced, were enrolled in 76 centers in Europe, Canada, Australia, Israel, Kuwait, Hong Kong and South Korea. Enrollment took place between February 21, 2006, and August 19, 2009.
4
The trial was conducted in accordance with the Helsinki Declaration. The Institution Review Board or Medical Ethic Committee approved the study protocol and all patients provided their informed consent. Inclusion criteria required that patients had been implanted with a dual-chamber device for a minimum duration of 2 years, received more than 40% ventricular pacing as detected by the device over a period of >4 weeks before replacement and that there was no clinical need to change the pacing mode or the AV intervals. Exclusion criteria are detailed elsewhere (see ref 13) and included cardiac resynchronization therapy indication, permanent AF, permanent 3rd degree AV Block, age less than 18 years, and neurocardiogenic syncope as primary pacemaker indication. Randomization was provided by an electronic database and was stratified by left ventricular ejection fraction (LVEF) - ≤40% or >40% - and by pacemaker versus ICD.
Follow-up and study end points Clinical information was retrieved at baseline and at scheduled follow-up visits at 1 month, 12 months and 24 months. After the 2 years follow-up visit, patients remained in the study and clinical and device data were collected until the last enrolled subject reached 24 months of follow-up. Daily total AT/AF duration as detected by the device was retrieved from device memory. The main objective of this secondary analysis was to identify predictors of persistent AT/AF among baseline patient characteristics and pacing modalities. The primary end point was persistent AT/AF, defined as either longer than 7 consecutive days with device diagnostics
5
showing 20 or more hours in AT/AF or AT/AF interrupted by atrial cardioversion or AT/AF present at 2 consecutive follow-up visits. This analysis was pre-specified in the study protocol and in the statistical analysis plan. We also measured other AT/AF-related end points such as the first occurrence of at least 5 minutes, 1 hour and 6 hours of AT/AF, and permanent AT/AF defined as one of the following: 7 consecutive days with device diagnostics showing AT/AF for ≥20 hours and either failed cardioversion or the investigator decided not to cardiovert the patient. Device Programming Once a patient was randomized, the following programming requirements were defined: patients in the DDD arm should be programmed as in the exchanged device, without a change in sensed and paced AV intervals, and no features switched on which had not been available or switched on in the previous device, while patients in the MVP arm should be programmed in AAI(R)<=>DDD(R) pacing mode, with Sensed AV interval, Paced AV interval and Rate adaptive AV left to physician discretion. Mode switch and AT/AF detection were required to be enabled in both study arms.
Medical Treatment Medical treatment was left to physician discretion and investigators were requested to treat HF patients with the optimal medical therapy available as described in the ESC Guidelines for the diagnosis and treatment of chronic HF (16). Statistical analysis Descriptive statistics were reported as mean and standard deviation for normally distributed continuous variables, or median with 25th-75th percentile interquartile range (IQR) in the case of skewed distribution. The incidence of end points at 2 years follow-up was estimated by the
6
Kaplan–Meier method and compared with the log-rank test. Cox regression was used to calculate hazard ratio (HR) with 95% confidence interval (CI). Univariable and multivariable Cox regression analysis was used to identify risk factors for persistent AT/AF occurrence. Two Cox regression models were estimated, one using the median PR interval and one using the upper quartile PR interval value. Beside baseline patient’s characteristics, also ventricular pacing percentage, as retrieved from device memory, was tested as a predictor of persistent AT/AF. We chose to use ventricular pacing percentage as estimated in the first 3 months of each patient observation period, in order to characterize the patient with a variable measured in a follow-up period as near as possible to baseline evaluation. Patients were deemed to have high ventricular pacing percentage if their ventricular pacing percentage was equal or higher than 10% and were compared with patients with ventricular pacing percentage lower than 10%. The main analysis considered all occurrences of persistent AT/AF. A sensitivity analysis was also performed excluding persistent AT/AF events which occurred in the first 3 months, to account for the fact that baseline ventricular pacing was estimated in the first 3 follow-up months after implant. The 2 year incidence of persistent AT/AF was estimated also in patients sub-groups selected as a function of PR interval, ventricular pacing percentage and randomization arm. Statistical tests were two-tailed, and p<0.05 was considered significant. The analysis adhered to the intent-to-treat principle. All the statistical analyses were performed using SAS 9.3. Results All 605 patients randomized in the PreFER MVP study were included in the analysis. Baseline patient characteristics are described in table I. In total, 186 (30.7%) patients had a known history of AF and 43 (7.1%) patients had a history of atrial tachycardia or atrial flutter.
7
Mean follow-up was 2.2 ± 1.0 years. Clinically defined persistent AT/AF Clinically defined persistent AT/AF was observed in 71 patients (actuarial incidence: 13.2%, CI=10.6%-16.5%) at 2 years. In univariable Cox regression analysis, persistent AT/AF was significantly associated with prior AF, coronary artery disease, cardiomyopathy, use of antiarrhythmic drugs, LVEF, ventricular pacing percentage and PR interval, but not with the randomization group, as shown in table II. In particular, the risk of developing persistent AT/AF for the whole study population increased by a factor 4.09 (HR=4.09, CI=2.11-7.94, p<0.001) in patients with ventricular pacing percentage higher than 10%. These patients represent two third of the whole patient population as shown in the ventricular pacing percentage distribution (figure 1). Median ventricular pacing percentage was 53.4% (lower-upper quartile 2.7%-96.0%) in the whole population, 88.0% (lower-upper quartile 44.5%-98.4%) in the DDD arm and 5.1% (lowerupper quartile 0.5%-64.2%) in the MVP arm. The association between persistent AT/AF and baseline PR interval was found for PR intervals longer or equal to 230 ms, corresponding to the upper quartile PR value in the studied population (HR =1.84, CI=1.00-3.37, p=0.049) while it was not significant (HR =1.35, CI=0.71-2.56, p=0.355) for PR≥210 ms, corresponding to the median PR interval in the studied population. Figure 2 shows time to first persistent AT/AF for the entire population in patients with ventricular pacing percentage ≤10% (continuous line) and >10% (dotted line). Figure 3 shows time to first persistent AT/AF in patients with PR interval < 230 ms (continuous line) and ≥ 230 ms (dotted line). Multivariable Cox regression analysis (table II) demonstrated that only a history of prior AT/AF (HR =2.85, CI=1.20-6.22, p=0.017) and high ventricular pacing percentage (HR =3.24, CI=1.13-
8
9.31, p=0.029) were independent predictors for persistent AT/AF. No qualitative differences were identified by repeating Cox Regression model on the sensitivity analyses. Secondary end points In the first 2 years post randomization 44.3% of the patients experienced 1 or more days with at least 5 minutes of device-detected AT/AF, 35.7% with at least 1 hour of AT/AF, 18.8% with at least 6 hours of AT/AF, while 7.9% of the patients had AT/AF longer than 7 days and 3.5% of patients were considered having permanent AT/AF. The randomization arm was not associated with statistically significant differences for any of the described AT/AF conditions when evaluating the whole study population. In the subgroup of patients with long PR interval (PR≥230 ms) MVP was related to a higher incidence of persistent AT/AF when compared with DDD pacing (HR 3.41, CI=1.10-10.6, p=0.024). In the subgroup of patients with shorter PR interval (PR<230 ms), patients randomized to MVP mode and having a ventricular pacing percentage lower than the median pacing percentage were associated with lower incidence of persistent AT/AF either when compared with patients randomized to DDD pacing and having a ventricular pacing percentage lower than the median pacing percentage (HR 0.25, CI=0.067-0.900, p=0.034) or when compared with patients randomized to MVP pacing and having a ventricular pacing percentage higher than the median pacing percentage. In patients randomized to the MVP arm, the median (IQR) ventricular pacing percentage was 50.5 % (10.9%-97.2%) in patients who developed persistent AT/AF and 2.7% (0.4%47.8%) in patients who did not develop persistent AT/AF.Discussion Main results The main result of this secondary analysis of PreFER MVP study was that in patients referred for replacement of an implanted pacemaker or ICD, prior AT/AF and high ventricular pacing 9
percentage were independent predictors of persistent AT/AF during the follow-up. In patients with long PR intervals (PR≥230 ms) MVP mode was associated with higher risk of persistent AT/AF compared with DDD mode. In patients with shorter PR intervals (PR<230 ms) MVP mode was associated with lower risk of persistent AT/AF compared with DDD mode, in patients with a low percentage of ventricular pacing. Atrial fibrillation prevalence in patients wearing implantable cardiac devices Our data confirm previous findings about the high prevalence of AT/AF in pacemaker and defibrillator patients [1-3]. A history of AT/AF was present in 37.8% of patients at baseline and these arrhythmias were documented in the device memories in 44.3% of patients during the 2year observation period; in real clinical practice AT/AF prevalence has to be estimated even higher when including permanent AT/AF which was an exclusion criterion in our study. Predictors of persistent atrial fibrillation In our patient population the development of persistent AT/AF was significantly associated with high ventricular pacing percentage (table II and figure 2). This finding confirms results of previous studies (4-7, 9). Some small clinical studies (4-6) and the post-hoc analyses of the MOST trial (7) suggested the association between the risk to develop AT/AF and high ventricular pacing percentage. The SAVEPACe trial (9), in 1065 patients with sinus-node disease, intact atrioventricular conduction, and normal QRS interval, compared conventional dual-chamber pacing with dual-chamber minimal ventricular pacing using pacemaker features designed to promote spontaneous atrioventricular conduction and to minimize ventricular pacing. The results showed that the incidence of persistent AT/AF was significantly reduced by 40% (p=0.009) in the group of patients assigned to minimal ventricular pacing compared with the DDD arm, in which the mean right ventricular pacing was 99%.
10
In our study, persistent AT/AF was not reduced in the MVP arm of the PreFER MVP trial, despite the fact that the MVP mode significantly reduced ventricular pacing percentage compared with DDD mode. This represents an intriguing issue, for which we propose the following possible explanations. First, PreFER MVP (15), by study design, selected a population of patients who were paced more than 40% of time by previous pacemaker or ICD and who survived their first implant without developing permanent AF and were therefore more predisposed to tolerate long term ventricular pacing. Secondly, risk for persistent AT/AF strongly increased for ventricular pacing percentage higher than 10%, and in the MVP arm of the study, in spite of a median value of 5.1%, a proportion of patients continued to be highly paced in the ventricle (upper quartile was 64.2%). Interestingly, in patients randomized to the MVP arm, the median ventricular pacing percentage was 50.5 % in patients who developed persistent AT/AF and 2.7% in patients who did not develop persistent AT/AF. Finally, persistent AT/AF was significantly associated with long PR (PR≥230 ms) and this finding may have overwhelmed the benefits of ventricular pacing reduction in MVP patients. As a matter of fact, in the DANPACE trial (10, 17), which randomly assigned 1415 sick sinus syndrome patients referred for first pacemaker implantation to AAIR (707) or DDDR (708) pacing, followed for a mean of 5 years, an excess of AF was observed in patients paced with single lead atrial pacing as compared with those paced in DDD mode, in particular in patients with prolonged intrinsic PR interval. Clinical implications As reported by Botto et al. (15), the PreFER MVP study aimed to compare MVP and DDD pacing modes in patients indicated to elective pacemaker or ICD replacement with high prior ventricular pacing percentage. The main analysis showed that in these patients, with clinically well tolerated long term exposure to more than 40% pacing in the ventricle, a strategy to
11
minimize ventricular pacing is not superior in reducing incidence of cardiovascular hospitalization compared to standard DDD pacing. Findings of the present analysis showing the association between ventricular pacing percentage greater than 10% and persistent AT/AF confirms the favorable role of pacing modalities to reduce unnecessary ventricular pacing in selected patients with normal spontaneous AV conduction. Reducing ventricular pacing below 10% may require dedicated algorithms and may be difficult to obtain by simply programming long AV interval. Indeed, subgroup analyses in our study showed that patients with long baseline PR interval (PR≥230 ms) had a higher incidence of persistent AT/AF if programmed in MVP compared with DDD pacing. We hypothesize that enabling very long intrinsic atrioventricular conduction times, as in the MVP arm, may cause AT/AF in patients with long PR intervals who were used to being chronically paced in the ventricle for a long time. As reported earlier (17-18), a long PR interval could be responsible for a higher incidence of AF, either in patients with or without history of AF. The actual reason of this association remains still hypothetical. Conversely, patients with shorter PR interval (PR<230 ms) randomized to MVP mode and having a ventricular pacing percentage lower than 5.1% (median pacing percentage in MVP mode) showed lower persistent AT/AF either compared with patients in DDD mode and low ventricular pacing percentage or patients in MVP mode and high ventricular pacing percentage. These data suggest that MVP is an effective pacing mode in patients with normal or only slightly prolonged PR interval, while pacing at a relatively short AV interval may be better in patients with a prolonged PR interval. When comparing MVP and DDD pacing, clinical benefit may derive from the balance of two conflicting factors, AV synchrony and ventricular pacing percentage. Although DDD pacing guarantees AV synchrony, unnecessary pacing in the right ventricle may cause ventricular remodeling due to the consequent change in electrical activation and contraction pattern of the
12
ventricles. Conversely, MVP preserves a normal ventricular contraction pattern, which is expected to reduce ventricular remodeling in comparison with a higher percentage of ventricular pacing, but at the cost of allowing a prolonged AV conduction in some patients. In patients undergoing pacemaker or ICD replacement with high percentage ventricular pacing history, the choice of the best pacing mode should be guided by careful evaluation of several clinical factors, including ventricular pacing percentage in the previous device and spontaneous PR interval. Our data suggest that programming the MVP mode is of major importance in reducing the percentage of ventricular pacing and hence the incidence of AF. Only in patients with a prolonged PR interval (≥ 230 ms) a DDD mode, with pacing at a relatively short AV interval, should be recommended. Study limitations We report results of secondary analyses of the PreFER MVP study. The statistical methods of the analyses on the main endpoint, persistent AT/AF, were pre-specified in the study protocol and in the statistical analysis plan, with the only exemption of the ventricular pacing percentage cut-off used to evaluate the association of this variable with persistent AT/AF. In particular, the cut-off (10%) used to divide the patient population between patients with low vs. high ventricular pacing percentage was chosen since it was the distribution tertile and possibly a good candidate to differentiate pacing modes devoted to minimize ventricular pacing from standard DDD pacing based on the observation of the skewed distribution of ventricular pacing percentage as measured in our study (figure 1) and in previous trials. (11-12, 19-20) The primary end point, persistent AT/AF, consists of three components, new-onset persistent AT/AF in patients without previous history of AT/AF, persistent AT/AF in patients with history of paroxysmal AT/AF, and recurrence of persistent AT/AF in patients with history of persistent
13
AT/AF. We cannot exclude that different mechanisms are responsible for the development of persistent AT/AF in these different situations. The use of antiarrhythmic drugs could have interfered with arrhythmia occurrence. The results of our analyses should be regarded as characteristics of the studied population of pacemaker and ICD replacement patients, with clinically well tolerated long term exposure to more than 40% pacing in the ventricle. In particular the results refer to conventional right atrial appendage pacing. We can not exclude that different results might have been observed in case of alternative site atrial pacing. Conclusions In pacemaker and ICD replacement patients, high percentage of ventricular pacing is associated with higher risk of persistent AT/AF. The use of algorithms which minimize unnecessary right ventricular pacing may benefit patients with normal spontaneous AV conduction but should be evaluated with caution in patients with long PR interval. Funding The PreFER MVP study was supported and sponsored by Medtronic Bakken Research Center. Acknowledgements The authors thank Bart Gerritse and Lidwien Vainer for contributing to and reviewing scientific and statistical methods and manuscript draft. Disclosures Renato Ricci has received consultant honoraries from Biotronik and Medtronic, and speaker fees from Biotronik. Giovanni Luca Botto has received research grants from Boston Scientific, St. Jude Medical, Bayer Healthcare, Gilead, Sanofi, Medtronic, consultant honoraries from Biotronik, Boston Scientific, St. Jude Medical, MSD, Bayer Healthcare, Sanofi, Medtronic, and speaker fees from Boston 14
Scientific, St. Jude Medical, Bayer Healthcare, Boheringer, Sanofi, Sorin, Pfizer, MSD and Medtronic. Juan M Bénézet has received consultant honoraries and speaker fees from Medtronic. Jens Cosedis Nielsen has received speaker fees from Biotronik and Biosense Webster, consultant honoraries from Boston Scientific, and research grants for the MANTRA-PAF trial from Biosense Webster. Luc De Roy received research and/or fellowship funds or consultant fees from Medtronic, St Jude, Boston, Sorin and Biotronic. Olivier Piot has received consultant honoraries from St Jude Medical, Sorin, Biotronik and Medtronic.
Aurelio Quesada has received research grants from Sorin
and Medtronic. Diego Vaccari has received research grants from St. Jude Medical, and speaker fees from Biotronik. Milan Kozak received consultant honoraries from Boston Scientific, Medtronic and Biotronik. Lorenza Mangoni and Andrea Grammatico are employee of the Medtronic EMEA Regional Clinical Centre. References 1. Israel CW, Grönefeld G, Ehrlich JR, Li YG, Hohnloser SH. Long-term risk of recurrent atrial fibrillation as documented by an implantable monitoring device: implications for optimal patient care. J Am Coll Cardiol. 2004;43:47-52. 2. Healey JS, Connolly SJ, Gold MR et al. ASSERT Investigators. Subclinical atrial fibrillation and the risk of stroke. N Engl J Med. 2012;366:120-9. 3. Fuster V, Rydén LE, Cannom DS et al. American College of Cardiology Foundation/American Heart Association Task Force. 2011 ACCF/AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. Circulation. 2011;123:e269-367.
15
4. Andersen HR, Cosedis Nielsen J, Thomsen PE, Thuesen L, Mortensen PT, Vesterlund T, Pedersen AK. Long-term follow-up of patients from a randomised trial of atrial versus ventricular pacing for sick-sinus syndrome. Lancet 1997; 350:1210–6. 5. Cosedis Nielsen J, Kristensen L, Andersen HR, Mortensen PT, Pedersen OL, Pedersen AK. A randomized comparison of atrial and dual chamber pacing in 177 consecutive patients with sick sinus syndrome: echocardiographic and clinical outcome. J Am Coll Cardiol 2003; 42:614–23. 6. Kristensen L, Cosedis Nielsen J, Mortensen PT, Pedersen OL, Pedersen AK, Andersen HR. Incidence of atrial fibrillation and thromboembolism in a randomised trial of atrial versus dual chamber pacing in 177 patients with sick sinus syndrome. Heart 2004; 90:661–6. 7. Sweeney MO, Hellkamp AS, Ellenbogen KA, Greenspon AJ, Freedman RA, Lee KL, Lamas GA; MOde Selection Trial Investigators. 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; 107:2932–7. 8. Gillis AM. Redefining physiologic pacing: lessons learned from recent clinical trials. Heart Rhythm 2006; 3:1367-72. 9. Sweeney MO, Bank AJ, Nsah E, Koullick M, Zeng QC, Hettrick D, Sheldon T, Lamas GA; Search AV Extension and Managed Ventricular Pacing for Promoting Atrioventricular Conduction (SAVE PACe) Trial. Minimizing Ventricular Pacing to Reduce Atrial Fibrillation in Sinus-Node Disease. N Engl J Med 2007; 357:1000-8.
16
10. Cosedis Nielsen J, Thomsen PE, Højberg S et al. DANPACE Investigators. A comparison of single-lead atrial pacing with dual-chamber pacing in sick sinus syndrome. Eur Heart J 2011; 32:686–96. 11. Sweeney MO, Shea JB, Fox V, Adler S, Nelson L, Mullen TJ, Belk P, Casavant D, Sheldon T. Randomized pilot study of a new atrial-based minimal ventricular pacing mode in dual-chamber implantable cardioverter-defibrillators. Heart Rhythm 2004; 1:160-7 12. Gillis AM, Pürerfellner H, Israel CW, Sunthorn H, Kacet S, Anelli-Monti M, Tang F, Young M, Boriani G; Medtronic Enrhythm Clinical Study Investigators. Reducing unnecessary right ventricular pacing with the managed ventricular pacing mode in patients with sinus node disease and AV block. Pacing Clin Electrophysiol 2006; 29:697705. 13. Quesada A, Botto G, Erdogan A, Kozak M, Lercher P, Cosedis Nielsen J, Piot O, Ricci R, Weiss C, Becker D, Wetzels G, De Roy L; PreFER MVP Investigators. Managed ventricular pacing vs. conventional dual-chamber pacing for elective replacements: the PreFER MVP study: clinical background, rationale, and design. Europace 2008; 10:321– 6. 14. Kurtz SM, Ochoa JA, Lau E, Shkolnikov Y, Pavri BB, Frisch D, Greenspon AJ. Implantation trends and patient profiles for pacemakers and implantable cardioverter defibrillators in the United States: 1993-2006. Pacing Clin Electrophysiol 2010; 33:70511. 15. Botto GL, Ricci RP, Bénézet JM, Cosedis Nielsen J, De Roy L, Piot O, Quesada A, Quaglione R, Vaccari D, Garutti C, Vainer L, Kozák M, on behalf of the PreFER MVP Investigators, Managed Ventricular Pacing compared to conventional dual-chamber pacing for elective replacement in chronically paced patients: results of the Prefer for 17
Elective Replacement MVP (PreFER MVP) randomized study. Heart Rhythm 2014; 11:992-1000. 16. Swedberg K, Cleland J, Dargie H et al. Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (updated 2005). Eur Heart J 2005; 26:1115-40. 17. Cosedis Nielsen J, Thomsen PE, Højberg S et al. DANPACE investigators. Atrial fibrillation in patients with sick sinus syndrome: the association with PQ-interval and percentage of ventricular pacing. Europace 2012; 14:682-9. 18. Cheng S, Keyes MJ, Larson MG, McCabe EL, Newton-Cheh C, Levy D, Benjamin EJ, Vasan RS, Wang TJ. Long-term outcomes in individuals with prolonged PR interval or first-degree atrioventricular block. JAMA 2009; 301:2571-7. 19. Almendral J, Arribas F, Wolpert C, Ricci R, Adragao P, Cobo E, Navarro X, Quesada A; DATAS Steering Committee; DATAS Writing Committee; DATAS Investigators. Dualchamber defibrillators reduce clinically significant adverse events compared with singlechamber devices: results from the DATAS (Dual chamber and Atrial Tachyarrhythmias Adverse events Study) trial. Europace. 2008;10(5):528-35 20. Barsheshet A, Moss AJ, McNitt S, Jons C, Glikson M, Klein HU, Huang DT, Steinberg JS, Brown MW, Zareba W, Goldenberg I; MADIT-II Executive Committee. Long-term implications of cumulative right ventricular pacing among patients with an implantable cardioverter-defibrillator. Heart Rhythm. 2011;8(2):212-8. Clinical perspectives This secondary analysis of the PreFER MVP clinical trial provides new evidence about the fact that in patients undergoing pacemaker or ICD replacement, high percentage of
18
ventricular pacing is associated with higher risk of developing persistent AT/AF. Our data suggest that the use of algorithms minimizing unnecessary right ventricular pacing is warranted in patients who have normal spontaneous AV conduction while it should be evaluated with caution in patients with long PR interval. These finding have clinical importance because patients undergoing pacemaker or ICD replacement who have a high percentage ventricular pacing history, represent a relevant portion of total implants. So far no data have been published about the choice of the best pacing mode in this specific patientpopulation.
19
Table I: Baseline patients characteristics
Subject Characteristics Male gender N (%) Age (years) Implanted device - Pacemaker (N,%) Implanted device - ICD (N,%)
Total (N = 605) 365 (60.3%) 75 ± 11 556 (91.9%) 49 (8.1%)
Primary Device (pacemaker only) Indication Sinus Node Dysfunction N (%)
372 (61.5%)
AV block N (%)
139 (23.0%)
Other brady indication N (%) Previous device implant duration (years) - mean ± SD Ventricular pacing percentage in the period preceding study enrollment - median (IQR)
45 (7.4%) 7.7 + 3.3 95% (IQR 75% – 99%)
History of Atrial fibrillation N (%)
186 (30.7%)
Paroxysmal N (%)
168 (27.8%)
Persistent N (%)
18 (3.1%)
History of Atrial flutter/tachycardia N (%)
43 (7.1%)
Cardiomyopathy N (%)
118 (19.5%)
Coronary Artery Disease N (%)
140 (23.1%)
Hypertension N (%)
307 (50.7%)
Diabetes mellitus N (%)
101 (16.7%)
Heart Failure N (%)
434 (71.7%)
NYHA Class I N (%)
186 (30.7%)
NYHA Class II N (%)
216 (35.7%)
NYHA Class III-IV N (%) LVEF (%) – mean ± SD Intrinsic P-R Interval (ms) median (IQR)
32 (5.3%) 53 ± 13 210 (170-230)
Cardiovascular Medications N (%) Aspirin/Acetylsalicylic acid N (%)
296 (48.9%)
Anticoagulants N (%)
147 (24.3%)
Diuretics N (%)
196 (32.4%)
Angiotensin-converting enzyme inhibitors N (%)
214 (35.4%)
Angiotensin II receptor blockers N (%)
112 (18.5%)
Beta-blockers N (%)
245 (40.5%)
Calcium antagonists N (%)
117 (19.3%)
Digitalis/digoxin N (%) Amiodarone N (%)
56 (9.3%) 100 (16.5%)
LVEF=left ventricle ejection fraction, NYHA=New York Heart Association, SD=standard deviation, IQR=25th-75th interquartile range
20
Table II Cox regression model for predictors of persistent AT/AF
Univariate
Multivariate
p value
Hazard Ratio
95% Hazard Ratio Confidence Limits
p value
Baseline Characteristic
Hazard Ratio
95% Hazard Ratio Confidence Limits
Prior AT/AF
3.52
2.30-5.38
<0.001
2.85
1.20-6.22
0.017
Coronary artery disease
1.95
1.26-3.01
0.003
1.19
0.21-6.76
0.848
Cardiomyopathy
2.20
1.41-3.41
<0.001
1.43
0.46-4.40
0.533
Antiarrhythmic drugs
2.50
1.59-3.94
<0.001
1.66
0.68-4.09
0.267
LVEF (%)
0.97
0.97-0.99
0.001
0.99
0.95-1.03
0.557
PR interval (≥230 ms)
1.84
1.00-3.37
0.049
2.11
0.87-5.10
0.097
1.35
0.88-2.05
0.167
2.43
0.95-6.22
0.063
4.09
2.11-7.94
<0.001
3.24
1.13-9.31
0.029
Randomization [MVP ON / MVP OFF] Ventricular pacing percentage (estimated in the first 3 months – cut off ≥10%)
Figure legend Figure 1 Distribution of ventricular pacing percentage as measured in the first 3 months of the observation period in the whole population
21
Figure 2 Time to first persistent AT/AF in patients with ventricular pacing percentage ≤ 10% (continuous line) and > 10% (dotted line) in the whole population, for the subgroup of 549 patients with complete data about ventricular pacing percentage. Figure 3 Time to first persistent AT/AF in patients with PR interval < 230 ms (continuous line) and ≥ 230 ms (dotted line) in the whole population, for the subgroup of 353 patients with complete data about PR interval.
22
Figure 1
Figure 2
Figure 3