Permanent Pacemaker Implantation Following Transcatheter Aortic Valve Implantation: Impact on Late Clinical Outcomes and Left Ventricular Function Marina Urena, John G. Webb, Corrado Tamburino, Antonio J. Muñoz-García, Asim Cheema, Antonio E. Dager, Vicenç Serra, Ignacio Amat-Santos, Marco Barbanti, Sebastiano Immè, Juan H. Alonso Briales, Luis Miguel Benitez, Hatim Al Lawati, Angela Maria Cucalon, Bruno García del Blanco, Javier López, Eric Dumont, Robert DeLarochellière, Henrique B. Ribeiro, Luis Nombela-Franco, François Philippon and Josep Rodés-Cabau Circulation. published online December 26, 2013; Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 2013 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circ.ahajournals.org/content/early/2013/12/26/CIRCULATIONAHA.113.005479
Data Supplement (unedited) at: http://circ.ahajournals.org/content/suppl/2013/12/26/CIRCULATIONAHA.113.005479.DC1.html
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation is online at: http://circ.ahajournals.org//subscriptions/
Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
Permanent Pacemaker Implantation Following Transcatheter Aortic Valve Implantation: Impact on Late Clinical Outcomes and Left Ventricular Function Running title: Urena et al.; TAVI, Pacemaker and Late Outcomes Marina Urena, MD1; John G. Webb, MD2; Corrado Tamburino, MD3; Antonio J. Muñoz-García, MD, PhD4; Asim Cheema, MD5; Antonio E. Dager, MD6; Vicenç Serra, MD7; Ignacio Amat-Santos, MD8; Marco Barbanti, MD2,3; Sebastiano Immè, MD3; Juan H. Alonso Briales, MD4; Luis Miguel Benitez, MD6; Hatim Al Lawati, MD5; Angela Maria Cucalon, MD6; Bruno García del Blanco, MD7; Javier López, MD, PhD8; Eric Dumont, MD D1; Ro Robe Robert bert be rt Nombela-Franco, co, M MD D1; DeLarochellière, MD1; Henrique B. Ribeiro, MD1; Luis Nombela-Franc François Philippon, MD, FRCPC, FHRS1; Josep Rodés-Cabau, MD MD1 1
Quebec Quebec Queb ec Heart Hea e rtt & Lung Institute, Laval University, Univerrsiity t , Quebec City, Que Quebec, ueebe b c, Canada; 2St-Paul’s Hospital, Hos Ho spital spi al, l, University Univ iveers ersity of British Columbia, Vancouver, Vanc nccouver, British C Columbia, o um ol mbia, bia Canada; 3Ferrarotto 4 Ho Hospital, , University Univ Un iver iv e siity of er of Catania, C taani Ca nia, a, Italy; Ita taly ly; Ho Hospital Hosp s it sp i al C Clínico lííni nico co ddee Mála M Málaga, ála laga ga,, Un ga U Universidad iv ver ersi sida si dadd dee M da Málaga, álag ál aga, ag a 5 6 Spain; S ain; St Sp St-Michael’s t-M Micchaael’ss H Hospital, ospita pital, l, T Toronto orronto U University, niveerssity y, To Tor Toronto, ronnto, nto, O Ontario, ntarrio io,, Ca Can Canada; nada;; Cl nada Clìnica C ìnicca ddee Occidente O cccidente ci dee Cali, Caalii, Cali, Caalii, Colombia; Colom olom mbi b a;; 7Hos Hospital H ospittal Universitari Unniverssit itaari ari Va V Vall ll dd’Hebron, ’H Hebrronn, n, U Universitat nive v rssittat A Autonoma utonnoma 8 de B Barcelona, arce ar celo ce lona lo na,, Barc B Barcelona, arccel eloona, ona, S Spain; paain in;; Ho Hospital H osp sppit ital al C Clinico lini li n co ni co U Universitario nive ni vers ve rsit rs i ar it ario io ddee Va Vall Valladolid, llad adol ad ollid id,, Va Val Valladolid, llad llad dol olid id,, Sp id Spai Spain an ai Addr Ad Address dres esss for for Correspondence: C rr Co rres espo pond nden ence ce:: Josep Rodés-Cabau, MD Quebec Heart Lung Institute, Laval University 2725 Chemin Ste-Foy Quebec City, Quebec, G1V 4G5 Canada Tel: 418-6568711 Fax: 418-6564544 E-mail: [email protected]
Journal Subject Codes: Treatment:[23] Catheter-based coronary and valvular interventions:other, Treatment:[23] Catheter-based coronary and valvular interventions:other, Etiology:[5] Arrhythmias, clinical electrophysiology, drugs, Etiology:[5] Arrhythmias, clinical electrophysiology, drugs
1 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
Abstract
Background—Very few data exist on the clinical impact of permanent pacemaker implantation (PPI) following transcatheter aortic valve implantation (TAVI). The objective of this study was to assess the impact of PPI following TAVI on late outcomes in a large cohort of patients. Methods and Results—A total of 1,556 consecutive patients without prior PPI undergoing TAVI were included. 239 patients (15.4%) required a PPI within the first 30 days following TAVI. At a mean follow-up of 22±17 months, no association was observed between the need for 30-day PPI and all-cause mortality (HR: 0.98, 95% CI:0.74-1.30, P=0.871), cardiovascular mortality (HR: 0.81, 95% CI: 0.56-1.17, P=0.270), and all-cause mortality orr rehospitalization due due to to heart hear he artt ar failure (HR: 1.00, 95% CI:0.77-1.30, P=0.980). A lower rate of sudden or unknown death was observed CI: ob bse serv rv ved iin n ppatients atiien entts ts with PPI (HR: 0.31, 95% C I: 00.11-0.85, I: .11-0.85, P=0.023). P=0 0.0 . 233). ) Patients Pat a ients with new PPI showed poorer ventricular (LVEF) over howed ow w a poore rerr eevolution vollutio on of lleft eftt ve ef vent ntri ricculaar eejection jecctioon ffraction raact ctiion (L LVEF) VEF)) ov ver ti ver time me ((P=0.017), P=0 =0.0 . 17 17), ), aand nd new to12-month new PP PPII wa wass an independent indep pen nde dent ntt ppredictor reedi dict ctor ct orr ooff LV LVEF EF ddecrease eccreeasse at at 66-- to o122-m - on onth th ffollow-up ollo ol l w-u lo w-up up ((estimated essti tim mateed mate coefficient: -2.26, CI:-4.07 -0.44, P= 0.013, -2. 2.26 2 , 95 26 95% % CI CI:: 4. :4.07 077 ttoo -0 0.444, 4 P = 0. 0.01 013, 01 3 R2: 0. 3, 00.121). 121) 12 1).. 1) Conclusions—The need for PPI was a frequent complication of TAVI, but it was not associated with any increase in overall or cardiovascular death or rehospitalization due to heart failure after a mean follow-up of ~ 2 years. Indeed, 30-day PPI was a protective factor for the occurrence of unexpected (sudden or unknown) death. However, new PPI did have a negative effect on left ventricular function over time.
Key words: transcatheter aortic valve implantation, pacemaker, aortic stenosis, percutaneous aortic valve replacement, sudden cardiac death
2 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
Introduction Transcatheter aortic valve implantation (TAVI) has become the treatment of choice for patients with aortic stenosis who are considered to be non-operable and a good alternative for those at high surgical risk.1 However, the occurrence of some periprocedural complications remains a concern. The need for permanent pacemaker implantation (PPI) following the procedure is one of the most frequent complications associated with TAVI, with an overall incidence of about 15% (~25% and 7% following TAVI with self-expandable (SEV) and balloon-expandable valves (BEV), respectively).1 Strong evidence supports the potential negative impact of right ventricular apical pacing, which has been associated with an increased rate of the combined endpoint of mort m mortality orttal alit ityy an it andd rehospitalization ehospitalization due to heart failure in patients with left ventricular dysfunction,2-4ventricular 5 6 ach hya yarrrhy rrhy hyth thm th mias ass,5, and pacing-induced cardiomyopathy cardiom omyyopathy in patients om pattie i ntss without with wi t out overt structural tachyarrhythmias,
hheart hear eart rt disease.7 H However, ow wev verr, ev eevidence ideence id encee oon n th thee cli clinical iniical im impact mpaact act off P PPI PI ffollowing ollo low winng T TAVI A I rremains AV em mai s sc main scar scarce a ce ar and an nd based base ba s d on ssmall se malll sstudies tu udiies w with ithh li it limi limited m te mi tedd (( 1 yyear) e r) follow-up. ea folllowloww-uup. up 8-111 Wh While Whil ilee thes tthese hesse st sstudies uddiees ddid idd no nott find fin any impact of of PPI P I on mortality, PP mor o ta tali liity,, concerns con once c rn rnss that that tthey hey m ay hhave avee be av been e uunderpowered en nder nd erpo er powe po w re we redd du duee to may inadequate sample size have been raised.12 Also, no studies to date have evaluated the impact of PPI on rehospitalizations due to heart failure, left ventricular function changes and sudden death. Finally, the vast majority of patients included in studies evaluating the impact of PPI following TAVI had received a SEV,8-11 and very few data exist on those patients receiving a BEV. The aims of this study were, therefore, to assess, in a large cohort of patients undergoing TAVI with BEV and SEV, the impact of new PPI on i) late outcomes (including mortality and rehospitalization due to heart failure) and ii) left ventricular function and functional status changes following the intervention.
3 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
Methods Study population A total of 1,811 consecutive patients who underwent TAVI with either BEV or SEV in 8 centers between January 2005 and February 2013 were screened. Of these, 233 patients were excluded due to pre-existing pacemaker implantation, and 22 patients due to an unsuccessful procedure without valve implantation. The final study population consisted of 1,556 patients (BEV: 858 patients; SEV: 698 patients). Patients were considered candidates for TAVI if they were at high or prohibitive predicted perioperative risk as evaluated by a heart team composed of cardiac surgeons and interventional cardiologists at each center. TAVI procedures were performed as previously described. descrrib ibed e .1 T ed The he study stu tudy y was conducted in accordance with the institutional ethics committee of each participating center, and an nd all all patients pati pa tien ti e ts provided en pro rovvided signed informed consen consent nt fo for or the procedur procedures. res e . Da Data ata were collected prospectively pros spe p ctively in eeach achh ccenter. ennter nter er.. Pr Pro Procedural oceeduuraal ccomplications om mpllicattions fo forr th the he pu pur purpose rpose ooff th rpo this iss sstudy tudyy w tudy were eree de er defi defined fine ine nedd according ac cco cord rdin rd i g to VARC-2 in VARC-2 C-2 criteria. crrit iteeria iaa.13 Indications ndications fo for or PPI PP PI In agreement with the ACC/AHA/HRS recommendations,PPI was indicated if third-degree or advanced second-degree atrio-ventricular block (AVB) at any anatomical level occurred and was not expected to resolve, or in the presence of sinus node dysfunction and documented symptomatic bradycardia.14 The indication of PPI in the presence of left bundle branch block (LBBB) with PR prolongation (>200 msec) not expected to normalize was at the discretion of the physician. The selection of a single-chamber or dual-chamber pacemaker was left to the implanter’s choice. Follow-up Follow-up was carried out through clinical outpatient visits and/or phone contacts at 30 days, 6
4 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
months, 12 months and yearly afterwards. No patient was lost during the follow-up period. Echocardiographic examinations at baseline were available in all patients, in 1279 patients at hospital discharge, and in 902 patients at 6- to 12-month follow-up (83% of patients alive at that point of time, 89% and 78% in the BEV and SEV groups, respectively, P=0.002). Left ventricular ejection fraction (LVEF) was calculated using the biplane modified Simpson’s method and left ventricular dysfunction was defined as LVEF 50%.15 Endpoints and definitions. The primary endpoint was defined as a composite of all-cause mortality and hospitalization due to heart failure at last follow-up. Secondary endpoints were: all cause-mortality, cardiovascular mortality, sudden cardiac death, composite of sudden cardiac death t and death of uunknown nkn wn nkno w ccause, ause au se, ehospitalization due to heart failure, functional class changes, and LVEF changes. Several sources rehospitalization off iinformation nfor nf orma or maati tion onn w eree used to investigate endpoints: er s: ooutpatient utpatient clinic cal a vvisits, issit itss, phone contacts with were clinical pati ien e ts, famili lies ess oorr ph phys sic icia ians ia ns aand nd rev rreview eview off m ediical re ecord cord ds to dete ddetermine eterm rmin inee ca auses uses ooff de dea ath wh whe en en patients, families physicians medical records causes death when 1 nece ne cess ce ssar ss a y. A ar ll eevents ll vennts nts we were re de efin efin ned ac aaccording cco co ord din ingg to o tthe he V he ARCAR C-22 cr Ccrit iteeria eria ia..13 Sudden Suudd den ccardiac arrdi diacc de death eath eath hw was a as necessary. All defined VARC-2 criteria.
defined as any ny uunexpected neexp xpec eccte ted de deat a h du dduee to ccardiac ardi ar diac di ac ddisease ise seas asse oc occu curr cu rrin rr in ng wi with thin th in 1 hhour o r aafter ou fter ft er ooff the onset oof death occurring within symptoms.16Death was classified as of unknown cause if the unexpected death failed to meet the confirmation criteria of sudden cardiac death, and the cause of death could not be determined after contact with the responsible physician or the patient’s family. Death of unknown cause was classified as cardiovascular death.13 Only readmissions with a primary diagnosis of heart failure at hospital discharge were considered as rehospitalizations due to heart failure. For patients with several hospitalizations due to heart failure, only the first episode was included in the analysis. Statistical analysis Qualitative variables are expressed as percentages and quantitative variables as mean ±standard
5 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
deviation or median (interquartile range) according to variable distribution, and compared using Chi-Square or Fisher exact test and sided t-test or Wilcoxon signed-rank test as appropriate. The primary composite endpoint and secondary endpoints were compared between PPI and no PPI and BEV and SEV groups using proportional hazard models (cumulative outcomes). All multivariate models were adjusted for baseline differences in the univariate analysis including variables with a p value 0.10. A landmark analysis with a landmark cut-off at 30 days was used to further investigate the impact of PPI on study outcomes. Thirty-day outcomes were assessed with a logistic regression model. Survival rates were summarized using Kaplan-Meier estimates and the log-rank test was used for comparison between groups. A linear general model for repeated epeated measures with interaction was used to compare the changes in LVEF at at different diff di f eren ff entt time time m points between PPI and no PPI groups. Further comparisons were performed using the Tukey’s technique. Predictors were ech hni niqque. que. P rediict ctoors or of LVEF changes over time me w ere analyzingg us uusing in ng a univariate and a multivariable linear model. The were considered with m ulltivariable lt li ineaar rregression egrress ress ssio io on mo ode del. l. T he rresults essultss w erre co con nsid i erred id d ssignificant igni ig n fica ni caantt w ithh pit pp-values valu valu lues ess 0 >0.1 .100 fo .1 for all). However, a higher rate of unexpected (sudden or unknown) death was observed in patients with no PPI and normal left ventricular function (P=0.043). Also, no negative impact of PPI was encountered in patients with at least moderate left ventricular dysfunction (LVEF 40%) (P>0.10 for all), with a protective effect on unexpected death in patients with normal or mildly depressed left ventricular function (P=0.023). Baseline clinical characteristics and procedural findings and clinical outcomes during the follow-up period according to the type of valve implanted are displayed in Online Data Supplement Table 2 and 3, respectively. Death or heart failure, death from any cause and from
8 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
cardiovascular causes, sudden cardiac death, sudden /unknown death, and hospitalizations due to heart failure were similar in the SEV and BEV groups (P>0.10 for all) The late outcomes according to the need for PPI following TAVI for the patients who had received a BEV or a SEV are shown in Table 7. There were no differences in any of the late outcomes between patients with and without PPI in each of the transcatheter valve type groups. In the SEV group, a trend towards a lower rate of sudden cardiac /unknown death was observed in patients with PPI (HR: 0.30, 95%CI: 0.09-1.02, P=0.053). In the BEV group, the risk of sudden cardiac/unknown death was similar in patients with and without PPI (HR: 0.28, 95% CI: 0.04-2.05, P=0.212). However, no significant interaction was found between the need of PPI and the he type of valve implanted for unexpected death (P ((P=0.997) =0.997) and sudden cardiac ddeath eaath ((P=0.984). P=0 =0.9 .9884)) .9 PPI, LVEF and functional status Changes are shown Ch Chan ange an gess in vvalve ge alve vee hemodynamics hemodynamics according to the he nneed eed for PPI ar re sh how ownn in Online Data Supplement Figure S up pp pplement Fig igu ure 1. 1. LV LVEF EF ssignificantly igni ig n fi ni f cantly iincreased nccreaaseed iin n overall oveeral ov eralll population po ula popu latioon on aatt 66-to to o 112-month 2-mo 2mont mo nthh follow-up (from 56±13% 59±11%, P
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circ.ahajournals.org/content/early/2013/12/26/CIRCULATIONAHA.113.005479
Data Supplement (unedited) at: http://circ.ahajournals.org/content/suppl/2013/12/26/CIRCULATIONAHA.113.005479.DC1.html
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation is online at: http://circ.ahajournals.org//subscriptions/
Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
Permanent Pacemaker Implantation Following Transcatheter Aortic Valve Implantation: Impact on Late Clinical Outcomes and Left Ventricular Function Running title: Urena et al.; TAVI, Pacemaker and Late Outcomes Marina Urena, MD1; John G. Webb, MD2; Corrado Tamburino, MD3; Antonio J. Muñoz-García, MD, PhD4; Asim Cheema, MD5; Antonio E. Dager, MD6; Vicenç Serra, MD7; Ignacio Amat-Santos, MD8; Marco Barbanti, MD2,3; Sebastiano Immè, MD3; Juan H. Alonso Briales, MD4; Luis Miguel Benitez, MD6; Hatim Al Lawati, MD5; Angela Maria Cucalon, MD6; Bruno García del Blanco, MD7; Javier López, MD, PhD8; Eric Dumont, MD D1; Ro Robe Robert bert be rt Nombela-Franco, co, M MD D1; DeLarochellière, MD1; Henrique B. Ribeiro, MD1; Luis Nombela-Franc François Philippon, MD, FRCPC, FHRS1; Josep Rodés-Cabau, MD MD1 1
Quebec Quebec Queb ec Heart Hea e rtt & Lung Institute, Laval University, Univerrsiity t , Quebec City, Que Quebec, ueebe b c, Canada; 2St-Paul’s Hospital, Hos Ho spital spi al, l, University Univ iveers ersity of British Columbia, Vancouver, Vanc nccouver, British C Columbia, o um ol mbia, bia Canada; 3Ferrarotto 4 Ho Hospital, , University Univ Un iver iv e siity of er of Catania, C taani Ca nia, a, Italy; Ita taly ly; Ho Hospital Hosp s it sp i al C Clínico lííni nico co ddee Mála M Málaga, ála laga ga,, Un ga U Universidad iv ver ersi sida si dadd dee M da Málaga, álag ál aga, ag a 5 6 Spain; S ain; St Sp St-Michael’s t-M Micchaael’ss H Hospital, ospita pital, l, T Toronto orronto U University, niveerssity y, To Tor Toronto, ronnto, nto, O Ontario, ntarrio io,, Ca Can Canada; nada;; Cl nada Clìnica C ìnicca ddee Occidente O cccidente ci dee Cali, Caalii, Cali, Caalii, Colombia; Colom olom mbi b a;; 7Hos Hospital H ospittal Universitari Unniverssit itaari ari Va V Vall ll dd’Hebron, ’H Hebrronn, n, U Universitat nive v rssittat A Autonoma utonnoma 8 de B Barcelona, arce ar celo ce lona lo na,, Barc B Barcelona, arccel eloona, ona, S Spain; paain in;; Ho Hospital H osp sppit ital al C Clinico lini li n co ni co U Universitario nive ni vers ve rsit rs i ar it ario io ddee Va Vall Valladolid, llad adol ad ollid id,, Va Val Valladolid, llad llad dol olid id,, Sp id Spai Spain an ai Addr Ad Address dres esss for for Correspondence: C rr Co rres espo pond nden ence ce:: Josep Rodés-Cabau, MD Quebec Heart Lung Institute, Laval University 2725 Chemin Ste-Foy Quebec City, Quebec, G1V 4G5 Canada Tel: 418-6568711 Fax: 418-6564544 E-mail: [email protected]
Journal Subject Codes: Treatment:[23] Catheter-based coronary and valvular interventions:other, Treatment:[23] Catheter-based coronary and valvular interventions:other, Etiology:[5] Arrhythmias, clinical electrophysiology, drugs, Etiology:[5] Arrhythmias, clinical electrophysiology, drugs
1 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
Abstract
Background—Very few data exist on the clinical impact of permanent pacemaker implantation (PPI) following transcatheter aortic valve implantation (TAVI). The objective of this study was to assess the impact of PPI following TAVI on late outcomes in a large cohort of patients. Methods and Results—A total of 1,556 consecutive patients without prior PPI undergoing TAVI were included. 239 patients (15.4%) required a PPI within the first 30 days following TAVI. At a mean follow-up of 22±17 months, no association was observed between the need for 30-day PPI and all-cause mortality (HR: 0.98, 95% CI:0.74-1.30, P=0.871), cardiovascular mortality (HR: 0.81, 95% CI: 0.56-1.17, P=0.270), and all-cause mortality orr rehospitalization due due to to heart hear he artt ar failure (HR: 1.00, 95% CI:0.77-1.30, P=0.980). A lower rate of sudden or unknown death was observed CI: ob bse serv rv ved iin n ppatients atiien entts ts with PPI (HR: 0.31, 95% C I: 00.11-0.85, I: .11-0.85, P=0.023). P=0 0.0 . 233). ) Patients Pat a ients with new PPI showed poorer ventricular (LVEF) over howed ow w a poore rerr eevolution vollutio on of lleft eftt ve ef vent ntri ricculaar eejection jecctioon ffraction raact ctiion (L LVEF) VEF)) ov ver ti ver time me ((P=0.017), P=0 =0.0 . 17 17), ), aand nd new to12-month new PP PPII wa wass an independent indep pen nde dent ntt ppredictor reedi dict ctor ct orr ooff LV LVEF EF ddecrease eccreeasse at at 66-- to o122-m - on onth th ffollow-up ollo ol l w-u lo w-up up ((estimated essti tim mateed mate coefficient: -2.26, CI:-4.07 -0.44, P= 0.013, -2. 2.26 2 , 95 26 95% % CI CI:: 4. :4.07 077 ttoo -0 0.444, 4 P = 0. 0.01 013, 01 3 R2: 0. 3, 00.121). 121) 12 1).. 1) Conclusions—The need for PPI was a frequent complication of TAVI, but it was not associated with any increase in overall or cardiovascular death or rehospitalization due to heart failure after a mean follow-up of ~ 2 years. Indeed, 30-day PPI was a protective factor for the occurrence of unexpected (sudden or unknown) death. However, new PPI did have a negative effect on left ventricular function over time.
Key words: transcatheter aortic valve implantation, pacemaker, aortic stenosis, percutaneous aortic valve replacement, sudden cardiac death
2 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
Introduction Transcatheter aortic valve implantation (TAVI) has become the treatment of choice for patients with aortic stenosis who are considered to be non-operable and a good alternative for those at high surgical risk.1 However, the occurrence of some periprocedural complications remains a concern. The need for permanent pacemaker implantation (PPI) following the procedure is one of the most frequent complications associated with TAVI, with an overall incidence of about 15% (~25% and 7% following TAVI with self-expandable (SEV) and balloon-expandable valves (BEV), respectively).1 Strong evidence supports the potential negative impact of right ventricular apical pacing, which has been associated with an increased rate of the combined endpoint of mort m mortality orttal alit ityy an it andd rehospitalization ehospitalization due to heart failure in patients with left ventricular dysfunction,2-4ventricular 5 6 ach hya yarrrhy rrhy hyth thm th mias ass,5, and pacing-induced cardiomyopathy cardiom omyyopathy in patients om pattie i ntss without with wi t out overt structural tachyarrhythmias,
hheart hear eart rt disease.7 H However, ow wev verr, ev eevidence ideence id encee oon n th thee cli clinical iniical im impact mpaact act off P PPI PI ffollowing ollo low winng T TAVI A I rremains AV em mai s sc main scar scarce a ce ar and an nd based base ba s d on ssmall se malll sstudies tu udiies w with ithh li it limi limited m te mi tedd (( 1 yyear) e r) follow-up. ea folllowloww-uup. up 8-111 Wh While Whil ilee thes tthese hesse st sstudies uddiees ddid idd no nott find fin any impact of of PPI P I on mortality, PP mor o ta tali liity,, concerns con once c rn rnss that that tthey hey m ay hhave avee be av been e uunderpowered en nder nd erpo er powe po w re we redd du duee to may inadequate sample size have been raised.12 Also, no studies to date have evaluated the impact of PPI on rehospitalizations due to heart failure, left ventricular function changes and sudden death. Finally, the vast majority of patients included in studies evaluating the impact of PPI following TAVI had received a SEV,8-11 and very few data exist on those patients receiving a BEV. The aims of this study were, therefore, to assess, in a large cohort of patients undergoing TAVI with BEV and SEV, the impact of new PPI on i) late outcomes (including mortality and rehospitalization due to heart failure) and ii) left ventricular function and functional status changes following the intervention.
3 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
Methods Study population A total of 1,811 consecutive patients who underwent TAVI with either BEV or SEV in 8 centers between January 2005 and February 2013 were screened. Of these, 233 patients were excluded due to pre-existing pacemaker implantation, and 22 patients due to an unsuccessful procedure without valve implantation. The final study population consisted of 1,556 patients (BEV: 858 patients; SEV: 698 patients). Patients were considered candidates for TAVI if they were at high or prohibitive predicted perioperative risk as evaluated by a heart team composed of cardiac surgeons and interventional cardiologists at each center. TAVI procedures were performed as previously described. descrrib ibed e .1 T ed The he study stu tudy y was conducted in accordance with the institutional ethics committee of each participating center, and an nd all all patients pati pa tien ti e ts provided en pro rovvided signed informed consen consent nt fo for or the procedur procedures. res e . Da Data ata were collected prospectively pros spe p ctively in eeach achh ccenter. ennter nter er.. Pr Pro Procedural oceeduuraal ccomplications om mpllicattions fo forr th the he pu pur purpose rpose ooff th rpo this iss sstudy tudyy w tudy were eree de er defi defined fine ine nedd according ac cco cord rdin rd i g to VARC-2 in VARC-2 C-2 criteria. crrit iteeria iaa.13 Indications ndications fo for or PPI PP PI In agreement with the ACC/AHA/HRS recommendations,PPI was indicated if third-degree or advanced second-degree atrio-ventricular block (AVB) at any anatomical level occurred and was not expected to resolve, or in the presence of sinus node dysfunction and documented symptomatic bradycardia.14 The indication of PPI in the presence of left bundle branch block (LBBB) with PR prolongation (>200 msec) not expected to normalize was at the discretion of the physician. The selection of a single-chamber or dual-chamber pacemaker was left to the implanter’s choice. Follow-up Follow-up was carried out through clinical outpatient visits and/or phone contacts at 30 days, 6
4 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
months, 12 months and yearly afterwards. No patient was lost during the follow-up period. Echocardiographic examinations at baseline were available in all patients, in 1279 patients at hospital discharge, and in 902 patients at 6- to 12-month follow-up (83% of patients alive at that point of time, 89% and 78% in the BEV and SEV groups, respectively, P=0.002). Left ventricular ejection fraction (LVEF) was calculated using the biplane modified Simpson’s method and left ventricular dysfunction was defined as LVEF 50%.15 Endpoints and definitions. The primary endpoint was defined as a composite of all-cause mortality and hospitalization due to heart failure at last follow-up. Secondary endpoints were: all cause-mortality, cardiovascular mortality, sudden cardiac death, composite of sudden cardiac death t and death of uunknown nkn wn nkno w ccause, ause au se, ehospitalization due to heart failure, functional class changes, and LVEF changes. Several sources rehospitalization off iinformation nfor nf orma or maati tion onn w eree used to investigate endpoints: er s: ooutpatient utpatient clinic cal a vvisits, issit itss, phone contacts with were clinical pati ien e ts, famili lies ess oorr ph phys sic icia ians ia ns aand nd rev rreview eview off m ediical re ecord cord ds to dete ddetermine eterm rmin inee ca auses uses ooff de dea ath wh whe en en patients, families physicians medical records causes death when 1 nece ne cess ce ssar ss a y. A ar ll eevents ll vennts nts we were re de efin efin ned ac aaccording cco co ord din ingg to o tthe he V he ARCAR C-22 cr Ccrit iteeria eria ia..13 Sudden Suudd den ccardiac arrdi diacc de death eath eath hw was a as necessary. All defined VARC-2 criteria.
defined as any ny uunexpected neexp xpec eccte ted de deat a h du dduee to ccardiac ardi ar diac di ac ddisease ise seas asse oc occu curr cu rrin rr in ng wi with thin th in 1 hhour o r aafter ou fter ft er ooff the onset oof death occurring within symptoms.16Death was classified as of unknown cause if the unexpected death failed to meet the confirmation criteria of sudden cardiac death, and the cause of death could not be determined after contact with the responsible physician or the patient’s family. Death of unknown cause was classified as cardiovascular death.13 Only readmissions with a primary diagnosis of heart failure at hospital discharge were considered as rehospitalizations due to heart failure. For patients with several hospitalizations due to heart failure, only the first episode was included in the analysis. Statistical analysis Qualitative variables are expressed as percentages and quantitative variables as mean ±standard
5 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
deviation or median (interquartile range) according to variable distribution, and compared using Chi-Square or Fisher exact test and sided t-test or Wilcoxon signed-rank test as appropriate. The primary composite endpoint and secondary endpoints were compared between PPI and no PPI and BEV and SEV groups using proportional hazard models (cumulative outcomes). All multivariate models were adjusted for baseline differences in the univariate analysis including variables with a p value 0.10. A landmark analysis with a landmark cut-off at 30 days was used to further investigate the impact of PPI on study outcomes. Thirty-day outcomes were assessed with a logistic regression model. Survival rates were summarized using Kaplan-Meier estimates and the log-rank test was used for comparison between groups. A linear general model for repeated epeated measures with interaction was used to compare the changes in LVEF at at different diff di f eren ff entt time time m points between PPI and no PPI groups. Further comparisons were performed using the Tukey’s technique. Predictors were ech hni niqque. que. P rediict ctoors or of LVEF changes over time me w ere analyzingg us uusing in ng a univariate and a multivariable linear model. The were considered with m ulltivariable lt li ineaar rregression egrress ress ssio io on mo ode del. l. T he rresults essultss w erre co con nsid i erred id d ssignificant igni ig n fica ni caantt w ithh pit pp-values valu valu lues ess 0 >0.1 .100 fo .1 for all). However, a higher rate of unexpected (sudden or unknown) death was observed in patients with no PPI and normal left ventricular function (P=0.043). Also, no negative impact of PPI was encountered in patients with at least moderate left ventricular dysfunction (LVEF 40%) (P>0.10 for all), with a protective effect on unexpected death in patients with normal or mildly depressed left ventricular function (P=0.023). Baseline clinical characteristics and procedural findings and clinical outcomes during the follow-up period according to the type of valve implanted are displayed in Online Data Supplement Table 2 and 3, respectively. Death or heart failure, death from any cause and from
8 Downloaded from http://circ.ahajournals.org/ at Universitaet Duesseldorf on January 19, 2014
DOI: 10.1161/CIRCULATIONAHA.113.005479
cardiovascular causes, sudden cardiac death, sudden /unknown death, and hospitalizations due to heart failure were similar in the SEV and BEV groups (P>0.10 for all) The late outcomes according to the need for PPI following TAVI for the patients who had received a BEV or a SEV are shown in Table 7. There were no differences in any of the late outcomes between patients with and without PPI in each of the transcatheter valve type groups. In the SEV group, a trend towards a lower rate of sudden cardiac /unknown death was observed in patients with PPI (HR: 0.30, 95%CI: 0.09-1.02, P=0.053). In the BEV group, the risk of sudden cardiac/unknown death was similar in patients with and without PPI (HR: 0.28, 95% CI: 0.04-2.05, P=0.212). However, no significant interaction was found between the need of PPI and the he type of valve implanted for unexpected death (P ((P=0.997) =0.997) and sudden cardiac ddeath eaath ((P=0.984). P=0 =0.9 .9884)) .9 PPI, LVEF and functional status Changes are shown Ch Chan ange an gess in vvalve ge alve vee hemodynamics hemodynamics according to the he nneed eed for PPI ar re sh how ownn in Online Data Supplement Figure S up pp pplement Fig igu ure 1. 1. LV LVEF EF ssignificantly igni ig n fi ni f cantly iincreased nccreaaseed iin n overall oveeral ov eralll population po ula popu latioon on aatt 66-to to o 112-month 2-mo 2mont mo nthh follow-up (from 56±13% 59±11%, P
