Clinical Impact of a new Left Bundle Branch Block following TAVI Implantation: One Year Results of the TAVIK Cohort

Gerhard Schymik 1,*, Panagiotis Tzamalis 1,*, Peter Bramlage 2, Martin Heimeshoff 3, Alexander Würth 4, Rainer Wondraschek 1, Bernd-Dieter Gonska 4, Herbert Posival 3, Claus Schmitt 1, Holger Schröfel 3, Armin Luik 1

1 Medical Clinic IV - Department of Cardiology, Municipal Hospital Karlsruhe, Academic Teaching Hospital of the University of Freiburg, Germany

2 Institut für Pharmakologie und präventive Medizin, Mahlow, Germany

3 Clinic for Cardiac Surgery Karlsruhe, Germany

4 Medical Clinic III - Department of Cardiology, Vincentius Hospital Karlsruhe, Germany

* Both authors contributed equally

Correspondence:

Dr. Gerhard Schymik, MD

Medical Clinic IV–Municipal Hospital Karlsruhe

Moltkestrasse 90, 76133 Karlsruhe, Germany

Tel: +49 721 9742960; Fax: +49 721 9742909 e-mail: [email protected]

Journal: Clin Res Cardiol (Original article)

Word count: 2,870 Abstract: 389 Tables: 3 (Suppl 1)

Figures: 7 References: 42 Version: 02.10.2014

1 ABSTRACT

Background: Compared with surgical aortic valve replacement, transcatheter aortic valve implantation (TAVI) is associated with a higher risk of developing a new conduction disorder that necessitates permanent pacemaker implantation (PM). The most frequently observed conduction disorder is left bundle branch block (LBBB), which impairs left ventricular function.

Objectives: The primary objective of this study was to assess the incidence and prognostic significance of persistent new-onset LBBB following TAVI. Factors predictive of persistent new-onset LBBB were also explored.

Methods: This study included a total of 793 patients who underwent TAVI between May 2008 and April 2012. Patients were divided into two groups, those with persistent new-onset LBBB and those without persistent new-onset LBBB. Follow-up was conducted within one year of TAVI.

Results: Persistent new-onset LBBB was observed in 31.1% (n =197) out of 634 eligible patients. At 30 days and one year post-TAVI, the all-cause mortality rate was higher in patients with persistent new-onset LBBB (6.1%, n =12 and 20.8%, n =41, respectively) than in patients without new-onset LBBB (3.3%, n =10 and 13.0%, n =57, respectively; p = 0.014 and p = 0.010 for the two time points). Multivariate regression analyses revealed, that persistent new-onset LBBB was an independent predictor of all-cause mortality at one year (HR 1.84, 95%CI 1.35-2.02). PM implantation was observed slightly more frequently in patients with persistent new-onset LBBB (14.2%) than in those without (9.4%; HR 1.60, 95%CI 0.96-2.67). Risk factors for pacemaker (PM) were baseline RBBB (HR 6.23, 95%CI 3.76-10.33), chronic atrial fibrillation (HR 1.75, 95%CI 1.10-2.56) and the Medtronic CoreValve implantation (HR 2.40, 95%CI 1.55-3.75). At one-year follow-up, the mean survival of patients with PM (81.2%) was slightly lower, but not significantly different from that of patients without PM (85.0%; p = 0.377). Upon multivariable logistic binary regression analysis Medtronic CoreValve was associated with an increase rate of persistent new-onset LBBB (HR 2.52, 95%CI 1.67-3.80) and PM implantation. Mortality during one year of follow-up, however, was neither increased in the total population (p=0.232), nor in a subgroup of those with LBBB in a comparison of Medtronic CoreValve and Edwards SAPIEN.

Conclusion: This study demonstrated that persistent new-onset LBBB was associated with increased mortality in patients undergoing TAVI. Compared with the Edwards SAPIEN valve, implantation of the Medtronic CoreValve resulted in a higher rate of both persistent new-onset LBBB and PM but not death.

Keywords: left bundle branch block, permanent pacemaker, CoreValve, Edwards SAPIEN, mortality

2 INTRODUCTION

Transcatheter aortic valve implantation (TAVI) has become the standard of care for patients with severe aortic stenosis who are considered to be either inoperable or at high-risk for surgical aortic valve replacement (SAVR) (1-3). Although the preliminary results are very promising, current available risk scoring algorithms need to be optimized for transcatheter procedures (4, 5). In addition a more detailed post-procedural risk assessment may improve further technical developments (6). Up to date, a major concern is the development of post-procedural conduction abnormalities which may lead into pacemaker (PM) implantation (7-10). Atrioventricular conduction abnormalities are also observed after SAVR, but they are more common after TAVI, particularly in patients implanted with the self- expandable Medtronic CoreValve (Medtronic Inc., Minneapolis, Minnesota, USA). Furthermore, TAVI-induced conduction disturbances appear to be associated with a higher frequency of PM than those induced by SAVR (11-13). For both procedures, conduction abnormalities are thought to be caused by injuries to the native aortic valve and the proximal region of the left ventricular outflow tract (8).

The most commonly reported TAVI-induced atrioventricular conduction disorder is the left bundle branch block (LBBB) (8, 13). To date, several studies have investigated the incidence and predictive factors of new-onset LBBB following implantation of either the Medtronic CoreValve or the Edwards SAPIEN valve (Edwards Lifesciences LLC, Irvine, CA, USA) (14-18). However, the prognostic significance of persistent new-onset LBBB after TAVI is yet to be clarified (17, 19-22). The aim of the present study was to evaluate the clinical short- and long-term outcome of patients with a TAVI- induced persistent LBBB.

METHODS

Patients who received either the Medtronic CoreValve or the Edwards SAPIEN valve prosthesis were consecutively enrolled. The data were collected from the prospective, open TAVI Karlsruhe registry that enrolled consecutive patients who underwent TAVI over a four-year period. The structure of the registry has been published elsewhere (23). In brief: Joint discussions between cardiologists and cardiac surgeons were performed to select the most appropriate technique for valve replacement. Patients were selected for TAVI based on the following two criteria: 1) logistic EuroSCORE of ≥ 15; 2) age ≥ 75 years, a logistic EuroSCORE of < 15 and additional comorbidities not reflected in the EuroSCORE. These were: a) previous open heart surgery; b) malignancy with a life expectancy greater than one year, liver cirrhosis, severe pulmonary disease with long-term provision of oxygen or a Karnofsky Performance index between 50 and 70; c) frailty; and d) porcelain aorta. TAVI was also considered for patients who denied SAVR. TAVI was deemed to be inappropriate if the native aortic valve annulus was unsuitable or if life expectancy and quality of life were seriously affected by

3 comorbidities, including malignancy with a life expectancy of less than one year, major stroke, dementia with disability, uncontrolled congestive heart failure or cardiogenic shock. Exclusion criteria for this analysis were as follows: (1) prior PM; (2) prior intraventricular conduction abnormalities such as complete right bundle branch block (RBBB) or complete LBBB; and (c) other factors, such as conversion to open heart surgery, death before the first post procedural ECG.

Patients who were included were divided into two groups, those with persistent new-onset LBBB and those without. The baseline characteristics of patients were documented, and ECGs were obtained 24 h prior to TAVI and at hospital discharge. ECG data were analysed by a cardiologist blinded to the clinical data. The diagnosis of intraventricular conduction disorders was based on the recommendations of the World Health Organizational/International Society and Federation for Cardiology Task Force (24). Persistent new-onset LBBB was defined as an LBBB that developed peri- or post-procedurally and persisted at discharge. The requirement for PM was determined according to the recommendation of the cardiac society guidelines (25). Pacemakers were implanted in case of eligibility either immediately or within the next 12 to 24 hours. Patients who developed an LBBB but either died before hospital discharge or required PM were included in the group of patients with persistent new-onset LBBB. The primary endpoint was all-cause mortality at 30 days and at one year. Follow-up was conducted by telephone or clinical visits within one year of TAVI.

Procedures and devices

TAVI was performed by a multidisciplinary team composed of an interventional cardiologist, a cardiac surgeon, and an anesthesiologist specialized in cardiac surgery, and the team was trained together with catheterization laboratory and operating room personnel. Pre-TAVI evaluation of patients included cardiac catheterization, angiographic computed tomography (CT) and a transesophageal echocardiogram (TEE). CT scans were carefully analyzed to determine the distance between the coronary arteries and the aortic valve annulus, and the diameter of the aorta and the iliofemoral vessels. The diameter of the native aortic valve annulus was measured using CT combined with the TEE long-axis view at the level of leaflet insertion. Patients were implanted with the Edwards SAPIEN or SAPIEN XT valves (transfemoral and transapical access), or the Medtronic CoreValve (transfemoral access only).

Statistical analysis

Categorical values were compared using the 2 or Fisher’s exact test as appropriate. Continuous variables were compared using the student’s t-test or Wilcoxon rank sum test as appropriate. Cumulative outcomes at one year follow-up were assessed using Kaplan-Meier estimates and compared using log-rank test. A p-value of < 0.05 was considered to denote a statistically significant

4 difference. All variables with p<0.20 in univariate Cox regression analysis were entered into a multivariate Cox-regression analysis by the enter method to determine the effect of new-onset LBBB, adjusted for the other potential factors of the end-point. The association of the baseline characteristics on the development of a LBBB as well as the association of the ECG-characteristics with a pacemaker implantation were assessed with the use of a binary logistic regression analysis. Respectively, all the variables with p-value<0.20 were entered in multiple binary logistic regression analysis by the enter method. Data analysis was conducted with SPSS version 20 (IBM, Chicago, IL, USA).

RESULTS

Study population

The analysis was conducted using data from patients in the Karlsruhe registry (n = 1,000) who underwent TAVI between May 2008 and April 2012. Of these, 366 were excluded for the following reasons: 1) previous PM (n = 132); 2) incomplete data-sets (n = 59), conversion to open heart surgery (n=4), other transcatheter valve types (n = 5), 3) non-specific QRS >120ms (n = 7) and 4) pre-existing ventricular conduction abnormalities (n = 159) (Figure 1). Thus, a total of 634 patients were available for this analysis, all of which had a one-year follow-up available (Table 1). The Medtronic CoreValve was implanted in 19.2% (n = 122) of patients and the Edwards SAPIEN valve in 80.8% (n = 512).

Persistent new-onset LBBB

For all subsequent analyses, patients were categorized into two groups, those with persistent new- onset LBBB (n = 197; 31.1%) and those without (n = 437). With the exception of gender (p=0.07) and valve-type (p<0.001) baseline characteristics of the two groups of patients were balanced (Table 1).

The proportion of patients with an Edwards SAPIEN valve was higher in the No-LBBB group (85.4%; n = 373) and the proportion of patients with a CoreValve was higher in the group with persistent new- onset LBBB (59.4%; n = 58) with a p-value < 0.001. Persistent new-onset LBBB occurred significantly more frequent with the Medtronic CoreValve (n = 58 out of 122; 47.5%) compared to the Edward SAPIEN valve (139 out of 512; 27.1%) (p < 0.001) (Table 1), HR 2.43, 95%CI 1.62-3.65). Both univariate and multivariate logistic binary regression analysis indicated that use of the Medtronic CoreValve was predictive of persistent new-onset LBBB (HR: 2.518, 95% CI: 1.668-3.799, p < 0.001; Figure 2).

While 97.7% of patients were still alive at 30 days in patients without any conduction abnormalities (10 patients died), the rate was significantly (p=0.014) reduced in patients with a persistent new-onset LBBB (93.9%; 12 patients died) (Figure 3, upper panel). This was confirmed at one year where

5 survival rates were 87.0% for patients without and 79.2% for patients with persistent new-onset LBBB (p=0.010) (Figure 3, lower panel).

Multivariate regression analyses revealed, that persistent new-onset LBBB was an independent predictor of all-cause mortality at one year (hazard ratio [HR] 1.835, 95% confidence interval [CI]: 1.345-2.018; p = 0.008) (Figure 4). Other independent predictors of all-cause mortality at one year were: renal failure (HR 2.907, 95%CI: 1.609-5.251; p < 0.001), a left ventricular ejection fraction of < 40% (HR 2.063, 95%CI: 1.193-3.568; p = 0.015), acute myocardial infarction during the 90 days prior to TAVI (HR 1.770, 95%CI: 1.058-2.961; p = 0.026) and mitral valve lesion or defect >II° (HR 1.699; 95%CI 1.000-2.887; p=0.046).

New-onset versus perexistent LBBB

In an exploratory analysis we checked mortality rates in patients with new-onset LBBB versus those who had a prexistent LBBB (n=197) before TAVI (n=81; see Figure 1). Patient characteristics for the two groups were not statistically different (Supplemental figure 1). Kaplan-Meier analysis showed a higher death rate with new-onset LBBB (79.2%) than pre-existent (87.7%), but the difference reached no statistical significance (p=0.099) (Figure 5).

Permanent pacemaker implantation

A permanent pacemaker (PM) was indicated for 10.8% (n = 69) of all patients. The incidence of PM was significantly more frequent in patients implanted with the Medtronic CoreValve (21.3%, n = 26/122) compared with those implanted with the Edwards SAPIEN valve (8.4%, n = 43/512; HR 2.95, 95%CI 1.73-5.04; p <0.001).

PM implantation was observed slightly more frequently in patients with persistent new-onset LBBB (14.2%, n = 28/197) than in those without (9.4%, n = 41/437; HR 1.60, 95%CI 0.96-2.67; p = 0.075). At one-year follow-up, the mean survival of patients with PM (81.2%) was slightly lower, but not significantly different from that of patients without PM (85.0%; p = 0.377; Figure 6).

Risk factors for PM were further examined using ECG data from patients who were initially excluded to the analysis because of pre-existing conduction abnormalities (QRS complex of > 120 ms with LBBB or RBBB) (Table 2). Thus, the incidence of PM was assessed in a total of 793 patients (Figure 1), including those patients of the cohort with pre-existing LBBB (n = 81) or RBBB (n = 78). PM was required in 14.4% (n = 114) of patients with new and pre-existing LBBB. As illustrated in Table 2, multivariate analyses led to the identification of the following factors that predisposed patients to the requirement for PM: baseline RBBB (43.2%, n = 35; HR 6.23, 95%CI 3.76-10.33, p < 0.001), chronic atrial fibrillation (21.3%, n = 43; HR 1.75, 95%CI 1.10-2.56; p = 0.017) and the Medtronic CoreValve (24.7%, n = 37; HR 2.40, 95%CI 1.55-3.75, p < 0.001).

6 Edwards SAPIEN vs. Medtronic CoreValve

With Medtronic CoreValve resulting in more new-onset LBBB and more pacemaker implantations we aimed to determine the prognostic significance of valve type chosen. As illustrated in Table 3, patients receiving either valve were largely comparable with only minor trends for a higher rate of diabetes mellitus in CoreValve patients (39.9 vs. 32.4; p=0.15), and trend for a higher rate of recent myocardial infarction (12.7 vs. 6.6%; p=0.06) and prior CABG (16.2 vs. 10.7%; p=0.12) in the Edwards SAPIEN groups. Mortality during one year of follow-up, however, was neither increased in the total population (p=0.232), nor in a subgroup of those with LBBB (p=0.968; Figure 7).

DISCUSSION

The present study demonstrates, that the incidence of a TAVI induced LBBB is not a rare phenomenon. The overall incidence of persistent new-onset LBBB was 31.1%, with a significant higher rate using the Medtronic CoreValve (47.5%; n = 58/122) compared with the Edwards SAPIEN valve (27.1%; n = 139/512). The higher rate of persistent new-onset LBBB with the CoreValve relative to the SAPIEN valve is consistent with the available literature, which suggests that TAVI- induced LBBB is experienced by 29% to 65% of patients who undergo TAVI with the Medtronic CoreValve, and by 12 to 18% of patients who receive the Edwards SAPIEN valve (14, 17, 19, 21, 26, 27). Other studies have reported that a peak of conduction abnormalities is observed during days 4 to 6 and 7 to 9 post-TAVI (28), and this is also in accordance with the present study in which PM implantation was performed during the first few days post-procedure until patient discharge. However, conduction abnormalities may resolve during the first year following TAVI, especially in patients implanted with the Edwards SAPIEN Valve (19, 29, 30). These findings suggest that the development of new conduction abnormalities is a consequence of the TAVI procedure, and may be associated with the healing process. Indeed, the proximity of the atrioventricular conduction system to the aortic valve annulus renders it susceptible to procedural injury, caused by either wire manoeuvres or valve expansion against the left ventricular outflow tract (31).

Clinical outcomes

The present study demonstrates that patients with persistent new-onset LBBB had a significant higher mortality rate in the short and long-term follow-up. However, the prognostic significance of TAVI- induced persistent new-onset LBBB is currently a controversial subject. While it is well-established that cardiac resynchronisation therapy (CRT) in patients with poor left ventricular function and a wide QRS complex reduces mortality and hospitalization rates (32-35), the impact of left ventricular

7 dyssynchrony in patients with valvular heart disease remains unclear. This is despite the high volume of publications on TAVI-induced conduction abnormalities (20, 22, 36). In the present study, both univariate and multivariate analyses indicated that persistent new-onset LBBB was an independent risk factor for mortality. The increase in mortality among patients with LBBB has been highlighted by recent electrophysiological studies. In particular, continuous right ventricular pacing in patients with dual chamber devices was identified to be an independent predictor of all-cause mortality (37), and it was also associated with a decline in the ejection fraction (38, 39). Therefore, although the present study could not distinguish between a cardiac or non-cardiac cause of death, it is reasonable to assume that TAVI-induced dyssynchrony in patients with persistent new-onset LBBB was a key factor that contributed to the mortality rate.

Reasons for discrepant effects of new-onset LBBB have been outlined recently such as the exclusion of patients with early permanent pacemaker implantation from other analyses (18, 20) and the definition for LBBB applied, which has changed more recently. While we did not exclude patients from our analysis based on early pacemaker implants, our rate of LBBB was compatible with the one reported by Testa (20), but less than reported by Houthuizen (18).

Relevant in this context is the observation that new-onset LBBB may resolve without treatment in up to 50% of patients (40). We were not able to address this aspect however, because most of our patients will not return to our hospital on a regular basis. It appears reasonable to assume that excluding patients with an early resolution of LBBB would only pronounce the impact of LBBB for the others one the background of our results.

The higher incidence of persistent new-onset LBBB following implantation of the Medtronic CoreValve relative to the Edwards SAPIEN valve may be explained by procedural factors such as the deeper insertion of Medtronic CoreValve (17). Given the relationship between persistent new-onset LBBB and all-cause mortality in the entire study population, it may be expected that the mortality rate would be higher among patients receiving the Medtronic CoreValve. However, counter-intuitively, the CoreValve prosthesis did not lead to a higher mortality rate than the SAPIEN valve. These data are in agreement with other studies, with the CoreValve being associated with an increased frequency of persistent new-onset LBBB and the requirement for PM, but not with an increase in the mortality rate (41). The results of this study indicate that persistent new-onset LBBB was not linked to an immediate increase in the rate of new conduction disorders that require PM. Perhaps a continuous monitoring could reveal paroxysmal higher degree heart rhythm disorders and may therefore be able to identify higher risk patients.

Study limitations

Patients undergoing TAVI over a four-year enrolment period were prospectively documented and this analysis retrospectively designed. Data on a cardiac or non-cardiac cause of death were not

8 documented; therefore, more studies are necessary to elucidate the most common reasons for death among patients with TAVI-induced LBBB. Furthermore, additional investigation is required to determine whether a secondary relationship exists between new-onset LBBB and major postoperative complications. Diagnosis of LBBB was performed in accordance with electrocardiographic guidelines (42) by a cardiologist who was blinded to patients’ clinical outcomes. Finally we are not able to assess the data by valve type because the patient cohort undergoing CoreValve implantation was too small.

Implications for clinical practice

New-onset LBBB is among the most frequent complications of TAVI. Experience with surgical aortic valve implantation associated LBBB and reports of LBBB associated increased mortality risk after TAVI has led to a generous provision of pacemakers. The continuing development of the TAVI device and of implantation techniques has lessened this risk however and recent reports have been inconsistent with respect to the implications of new-onset LBBB. Caution should still be exercised, to avoid this deleterious complication.

Conclusions

Persistent new-onset LBBB was observed in approximately 30% of patients who underwent TAVI. At short and long-term follow-up, all-cause mortality was higher in the group of patients with persistent new-onset LBBB. Compared with the Edwards SAPIEN valve, implantation of the Medtronic CoreValve resulted in a higher rate of both persistent new-onset LBBB and PM but not death.

FUNDING SOURCES

This study was conducted without any external financial support.

DISCLOSURES

Gerhard Schymik (GS) and Holger Schröfel are proctors and Peter Bramlage (PB) consultant for Edwards Lifesciences. No conflict of interest was declared by the other authors.

AUTHOR CONTRIBUTIONS

All authors except PB established, and conducted the registry. GS, Panagotis Tzamalis (PT) and PB designed the statistical approach, analysis and interpretation. GS, PT and PB drafted the first version of the manuscript, which all other authors revised for important intellectual content. All authors approved the final version of the manuscript to be submitted.

9 REFERENCES

1. Joint Task Force on the Management of Valvular Heart Disease of the European Society of C, European Association for Cardio-Thoracic S, Vahanian A, Alfieri O, Andreotti F, Antunes MJ, Baron-Esquivias G, Baumgartner H, et al. (2012) Guidelines on the management of valvular heart disease (version 2012). Eur Heart J 33: 2451-96. 2. Diemert P, Lange P, Greif M, Seiffert M, Conradi L, Massberg S, Blankenberg S, Reichenspurner H, et al. (2014) Edwards Sapien XT valve placement as treatment option for aortic regurgitation after transfemoral CoreValve implantation: a multicenter experience. Clin Res Cardiol 103: 183-90. 3. Haussig S, Schuler G, Linke A (2014) Worldwide TAVI registries: what have we learned? Clin Res Cardiol 103: 603-12. 4. Seiffert M, Sinning JM, Meyer A, Wilde S, Conradi L, Vasa-Nicotera M, Ghanem A, Kempfert J, et al. (2014) Development of a risk score for outcome after transcatheter aortic valve implantation. Clin Res Cardiol 103: 631-40. 5. Sedaghat A, Sinning JM, Vasa-Nicotera M, Ghanem A, Hammerstingl C, Grube E, Nickenig G, Werner N (2013) The revised EuroSCORE II for the prediction of mortality in patients undergoing transcatheter aortic valve implantation. Clin Res Cardiol 102: 821-9. 6. Schymik G, Heimeshoff M, Bramlage P, Wondraschek R, Suselbeck T, Gerhardus J, Luik A, Posival H, et al. (2014) Ruptures of the device landing zone in patients undergoing transcatheter aortic valve implantation: an analysis of TAVI Karlsruhe (TAVIK) patients. Clin Res Cardiol. 7. Tang GH, Lansman SL, Cohen M, Spielvogel D, Cuomo L, Ahmad H, Dutta T (2013) Transcatheter aortic valve replacement: current developments, ongoing issues, future outlook. Cardiol Rev 21: 55-76. 8. van der Boon RM, Nuis RJ, Van Mieghem NM, Jordaens L, Rodes-Cabau J, van Domburg RT, Serruys PW, Anderson RH, et al. (2012) New conduction abnormalities after TAVI-- frequency and causes. Nat Rev Cardiol 9: 454-63. 9. Lim GB (2014) Valvular disease: implications of new-onset conduction abnormalities after TAVI. Nat Rev Cardiol 11: 127. 10. Martinez-Selles M, Bramlage P, Thoenes M, Schymik G (2014) Clinical significance of conduction disturbances after aortic valve intervention: current evidence. Clin Res Cardiol. 11. Bagur R, Rodes-Cabau J, Gurvitch R, Dumont E, Velianou JL, Manazzoni J, Toggweiler S, Cheung A, et al. (2012) Need for permanent pacemaker as a complication of transcatheter aortic valve implantation and surgical aortic valve replacement in elderly patients with severe aortic stenosis and similar baseline electrocardiographic findings. JACC Cardiovasc Interv 5: 540-51.

10 12. Simms AD, Hogarth AJ, Hudson EA, Worsnop VL, Blackman DJ, O'Regan DJ, Tayebjee MH (2013) Ongoing requirement for pacing post-transcatheter aortic valve implantation and surgical aortic valve replacement. Interact Cardiovasc Thorac Surg 17: 328-33. 13. Fraccaro C, Napodano M, Tarantini G (2013) Conduction disorders in the setting of transcatheter aortic valve implantation: a clinical perspective. Catheter Cardiovasc Interv 81: 1217-23. 14. Tzikas A, Piazza N, Geleijnse ML, Van Mieghem N, Nuis RJ, Schultz C, van Geuns RJ, Galema TW, et al. (2010) Prosthesis-patient mismatch after transcatheter aortic valve implantation with the Medtronic CoreValve system in patients with aortic stenosis. Am J Cardiol 106: 255-60. 15. Baan J, Jr., Yong ZY, Koch KT, Henriques JP, Bouma BJ, Vis MM, Cocchieri R, Piek JJ, et al. (2010) Factors associated with cardiac conduction disorders and permanent pacemaker implantation after percutaneous aortic valve implantation with the CoreValve prosthesis. Am Heart J 159: 497-503. 16. Fraccaro C, Buja G, Tarantini G, Gasparetto V, Leoni L, Razzolini R, Corrado D, Bonato R, et al. (2011) Incidence, predictors, and outcome of conduction disorders after transcatheter self- expandable aortic valve implantation. Am J Cardiol 107: 747-54. 17. Franzoni I, Latib A, Maisano F, Costopoulos C, Testa L, Figini F, Giannini F, Basavarajaiah S, et al. (2013) Comparison of incidence and predictors of left bundle branch block after transcatheter aortic valve implantation using the CoreValve versus the Edwards valve. Am J Cardiol 112: 554-9. 18. Houthuizen P, Van Garsse LA, Poels TT, de Jaegere P, van der Boon RM, Swinkels BM, Ten Berg JM, van der Kley F, et al. (2012) Left bundle-branch block induced by transcatheter aortic valve implantation increases risk of death. Circulation 126: 720-8. 19. Urena M, Mok M, Serra V, Dumont E, Nombela-Franco L, DeLarochelliere R, Doyle D, Igual A, et al. (2012) Predictive factors and long-term clinical consequences of persistent left bundle branch block following transcatheter aortic valve implantation with a balloon-expandable valve. J Am Coll Cardiol 60: 1743-52. 20. Testa L, Latib A, De Marco F, De Carlo M, Agnifili M, Latini RA, Petronio AS, Ettori F, et al. (2013) Clinical impact of persistent left bundle-branch block after transcatheter aortic valve implantation with CoreValve Revalving System. Circulation 127: 1300-7. 21. Houthuizen P, van der Boon RM, Urena M, Van Mieghem N, Brueren GB, Poels TT, Van Garsse LA, Rodes-Cabau J, et al. (2014) Occurrence, fate and consequences of ventricular conduction abnormalities after transcatheter aortic valve implantation. EuroIntervention 9: 1142-50. 22. Nazif TM, Williams MR, Hahn RT, Kapadia S, Babaliaros V, Rodes-Cabau J, Szeto WY, Jilaihawi H, et al. (2014) Clinical implications of new-onset left bundle branch block after

11 transcatheter aortic valve replacement: analysis of the PARTNER experience. Eur Heart J 35: 1599-607. 23. Schymik G, Schröfel H, Schymik JS, Wondraschek R, Süselbeck T, Kiefer R, Balthasar V, Luik A, et al. (2012) Acute and late outcomes of Transcatheter Aortic Valve Implantation (TAVI) for the treatment of severe symptomatic aortic stenosis in patients at high- and low- surgical risk. J Interv Cardiol 25: 364-74. 24. Willems JL, Robles de Medina EO, Bernard R, Coumel P, Fisch C, Krikler D, Mazur NA, Meijler FL, et al. (1985) Criteria for intraventricular conduction disturbances and pre- excitation. World Health Organizational/International Society and Federation for Cardiology Task Force Ad Hoc. J Am Coll Cardiol 5: 1261-75. 25. Brignole M, Auricchio A, Baron-Esquivias G, Bordachar P, Boriani G, Breithardt OA, Cleland J, Deharo JC, et al. (2013) 2013 ESC Guidelines on cardiac pacing and cardiac resynchronization therapy: the Task Force on cardiac pacing and resynchronization therapy of the European Society of Cardiology (ESC). Developed in collaboration with the European Heart Rhythm Association (EHRA). Eur Heart J 34: 2281-329. 26. Roten L, Wenaweser P, Delacretaz E, Hellige G, Stortecky S, Tanner H, Pilgrim T, Kadner A, et al. (2010) Incidence and predictors of atrioventricular conduction impairment after transcatheter aortic valve implantation. Am J Cardiol 106: 1473-80. 27. Piazza N, Onuma Y, Jesserun E, Kint PP, Maugenest AM, Anderson RH, de Jaegere PP, Serruys PW (2008) Early and persistent intraventricular conduction abnormalities and requirements for pacemaking after percutaneous replacement of the aortic valve. JACC Cardiovasc Interv 1: 310-6. 28. Bjerre Thygesen J, Loh PH, Cholteesupachai J, Franzen O, Sondergaard L (2014) Reevaluation of the indications for permanent pacemaker implantation after transcatheter aortic valve implantation. J Invasive Cardiol 26: 94-9. 29. Gutierrez M, Rodes-Cabau J, Bagur R, Doyle D, DeLarochelliere R, Bergeron S, Lemieux J, Villeneuve J, et al. (2009) Electrocardiographic changes and clinical outcomes after transapical aortic valve implantation. Am Heart J 158: 302-8. 30. Godin M, Eltchaninoff H, Furuta A, Tron C, Anselme F, Bejar K, Sanchez-Giron C, Bauer F, et al. (2010) Frequency of conduction disturbances after transcatheter implantation of an Edwards Sapien aortic valve prosthesis. Am J Cardiol 106: 707-12. 31. Nuis RJ, Van Mieghem NM, Schultz CJ, Tzikas A, Van der Boon RM, Maugenest AM, Cheng J, Piazza N, et al. (2011) Timing and potential mechanisms of new conduction abnormalities during the implantation of the Medtronic CoreValve System in patients with aortic stenosis. Eur Heart J 32: 2067-74. 32. Enriquez A, Barrero R, Bittner A, Frangini P, Baeza M, Millapan I, Gonzalez R, Vergara I (2013) [Cardiac resynchronization therapy in patients with heart failure: a 10-year experience]. Rev Med Chil 141: 968-76. 12 33. Frigerio M, Lunati M, Pasqualucci D, Vargiu S, Foti G, Pedretti S, Vittori C, Cattafi G, et al. (2014) Left ventricular ejection fraction overcrossing 35% after one year of cardiac resynchronization therapy predicts long term survival and freedom from sudden cardiac death: single center observational experience. Int J Cardiol 172: 64-71. 34. Gold MR, Leman RB, Wold N, Sturdivant JL, Yu Y (2014) The Effect of Left Ventricular Electrical Delay on the Acute Hemodynamic Response with Cardiac Resynchronization Therapy. J Cardiovasc Electrophysiol. 35. Rickard J, Cheng A, Spragg D, Green A, Leff B, Tang W, Cantillon D, Baranowski B, et al. (2014) Survival in Octogenarians Undergoing Cardiac Resynchronization Therapy Compared to the General Population. Pacing Clin Electrophysiol. 36. Calvi V, Puzzangara E, Pruiti GP, Conti S, Di Grazia A, Ussia GP, Capodanno D, Tamburino C (2009) Early conduction disorders following percutaneous aortic valve replacement. Pacing Clin Electrophysiol 32 Suppl 1: S126-30. 37. Luca Botto G, Ricci RP, Benezet JM, Cosedis Nielsen J, De Roy L, Piot O, Quesada A, Quaglione R, et al. (2014) Managed ventricular pacing compared with conventional dual- chamber pacing for elective replacement in chronically paced patients: Results of the Prefer for Elective Replacement Managed Ventricular Pacing randomized study. Heart Rhythm. 38. Ahmed M, Gorcsan J, 3rd, Marek J, Ryo K, Haugaa K, D RL, Schwartzman D (2014) Right ventricular apical pacing-induced left ventricular dyssynchrony is associated with a subsequent decline in ejection fraction. Heart Rhythm 11: 602-8. 39. Akerstrom F, Arias MA, Pachon M, Jimenez-Lopez J, Puchol A, Julia-Calvo J (2013) The importance of avoiding unnecessary right ventricular pacing in clinical practice. World J Cardiol 5: 410-419. 40. Roten L, Meier B (2014) Left bundle branch block after transcatheter aortic valve implantation: still a matter of concern? JACC Cardiovasc Interv 7: 137-9. 41. van der Boon RM, Van Mieghem NM, Theuns DA, Nuis RJ, Nauta ST, Serruys PW, Jordaens L, van Domburg RT, et al. (2013) Pacemaker dependency after transcatheter aortic valve implantation with the self-expanding Medtronic CoreValve System. Int J Cardiol 168: 1269- 73. 42. Surawicz B, Childers R, Deal BJ, Gettes LS, Bailey JJ, Gorgels A, Hancock EW, Josephson M, et al. (2009) AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part III: intraventricular conduction disturbances: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology. Circulation 119: e235-40.

13 14 Table 1. Baseline and patients’ procedural characteristics

Tot No LBBB (n=437) LBBB p-value LBBB vs. no multivariate p-value* al (n=197) LBBB (n= 634 ) n n (%) or mean±SD n (%) or (%) mean±SD or me an± SD Male 243 178 (40.7) 65 (33.3) 0.07 0.08 (38. 3) Age in years 82. 81.83±5.37 82.43±5.64 0.2 0.33 02± 4.4 6 Comorbidities Diabetes mellitus 214 142 (32.5) 72 (36.5) 0.32 (33. 8) Peripheral artery disease 82 54 (12.4) 28 (14.2) 0.45 (12. 9) Carotid stenosis 106 71 (16.2) 35 (17.8) 0.65 (16. 7) COPD 79 57 (13) 22 (11.2) 0.60 (12.

15 5) Pulmonary Hypertension 123 89 (20.4) 34 (17.3) 0.39 (19. 4) Renal failure (Creatinine ≥2.2 mg/dl, %) 37 25 (5.7) 12 (6.1) 0.89 (5.8 ) Major neurological dysfunction 92 65 (14.9) 27 (13.7) 0.81 (14. 5) Coronary artery disease 357 241 (55.1) 116 (58.9) 0.39 (56. 3) Acute MI (<90days, %) 73 49 (11.2) 24 (12.2) 0.79 (11. 5) Critical perioperative state 13 12 (2.7) 1 (0.5) 0.74 (2.1 ) Previous CABG 97 62 (14.2) 34 (17.3) 0.34 (15. 1) Previous Valve Surgery 21 14 (3.2) 7 (3.6) 0.81 (3.3 ) Mitral valve lesion or defect (≥II°, %) 73 49 (11.2) 24 (12.2) 0.79 (11. 5) Ejection fraction in % 59. 59.42±13.28 59.01±12.8 0.71 29± 4 13. 14

16 Euro-SCORE in% 21. 21.34±15.5 22.54±16.6 0.39 71± 3 13. 14 Valves Edwards SAPIEN 512 373 (85.4) 139 (70.6) (80. 8) <0.001 <0.001 Medtronic CoreValve 122 64 (14.6) 58 (29.4) (19. 2) Legend: p-value*: p-value in the multivariate analysis; MI, myocardial infarction; CABG, coronary artery bypass graft; COPD, chronic obstructive pulmonary disease

17 Table 2. Multivariate analysis of baseline and procedural characteristics that predisposed patients to permanent pacemaker implantation

Primary PM Without primary PM p-value *p HR (95% CI) LBBB n, (%) 10 (12.8) 69 (87.2) 0.569 RBBB n, (%) 35 (43.2) 46 (56.8) p<0.001 p<0.001 6.23 (3.76-10.33) AV-Block I. n, (%) 14 (18.7) 61 (81.3) 0.196 0.31 Chronic atrial 43 (21.3) 159 (78.7) 0.017 0.023 1.75 (1.10-2.56) fibrillation n, (%) Medtronic Core Valve n, 37 (24.7) 113 (75.3) p<0.001 p<0.001 2.40 (1.55-3.75) (%) Legend: LBBB, left bundle branch block; RBBB, right bundle branch block; p-value*: p-value in the multivariate analysis

18 Table 3. Baseline and procedural characteristics of the study population stratified by the use of the valve system

ESV MCV (n=122) p-value (n=512) n (%) or n (%) or n (%) or mean±SD mean±SD mean±SD Male 194 (37.9) 49 (40.2) 0.68 Age in years 82.01±5.48 82.04±5.37 0.95 Comorbidities Diabetes mellitus 166 (32.4) 48 (39.9) 0.17 Peripheral artery disease 62 (12.1) 20 (16.4) 0.23 Carotid stenosis 82 (16) 24 (19.7) 0.35 COPD 66 (12.9) 13 (10.7) 0.65 Pulmonary Hypertension 96 (18.8) 27 (22.1) 0.44 Renal failure (Creatinine ≥2.2 30 (5.9) 7 (5.7) 1.0 mg/dl, %) Major neurological dysfunction 77 (15) 15 (12.3) 0.48 Coronary artery disease 286 (55.9) 71 (58.2) 0.69 Acute MI (<90days, %) 65 (12.7) 8 (6.6) 0.06 Critical perioperative state 12 (2.3) 1 (0.8) 0.48 Previous CABG 83 (16.2) 13 (10.7) 0.16 Previous Valve Surgery 17 (3.3) 4 (3.3) 1.0 Mitral valve lesion or defect (≥II°, %) 56 (10.9) 17 (13.9) 0.35 Ejection fraction in % 59.16±13.10 59.84±13.29 0.92 Euro-SCORE in% 21.80±15.77 21.36±16.25 0.99 Legend: p-value*: p-value in the multivariate analysis; MI, myocardial infarction; CABG, coronary artery bypass graft; COPD, chronic obstructive pulmonary disease

19 Figure 1. Study population

20 Figure 2. Predictors of new persistent LBBB in a multivariate regression analysis at one year after TAVI

21 Figure 3. Kaplan-Meier survival curves for all-cause mortality in patients with and without persistent new-onset left bundle branch block at 30 days (upper panel) and one year (lower panel) after TAVI

22 Figure 4. Independent predictors of all-cause mortality at one year after TAVI by multivariate analysis.

23 Figure 5. Kaplan-Meier survival curves for all-cause mortality in patients with and without persistent new-onset left bundle branch block at one year after TAVI

24 Figure 6. Kaplan-Meier survival curves according to permanent pacemaker implantation at one year after TAVI

25 Figure 7. Kaplan-Meier survival curves for all-cause mortality in the total patient population (upper panel) and in patients with persistent new-onset left bundle branch block at one year (lower panel)

26 Supplemental Table 1. Baseline and patients’ procedural characteristics in patients with pre-existing versus new-onset LBBB

Total (n=278) LBBB pre-existent (n=81) LBBB new-onset (n=197) p-value n (%) or mean±SD n (%) or mean±SD n (%) or mean±SD Male 98 (35.3) 33 (40.7) 65 (33) 0.27 Age in years 82.23±5.47 81.76±5.01 82.44±5.64 0.35 Comorbidities Diabetes mellitus 107 (38.5) 35 (43.2) 72 (36.5) 0.34 Peripheral artery disease 38 (13.7) 10 (12.3) 28 (14.2) 0.85 Carotid stenosis 50 (18) 15 (18.5) 35 (17.8) 0.86 COPD 28 (10.1) 6 (7.4) 22 (11.2) 0.39 Pulmonary Hypertension 43 (15.5) 9 (11.1) 34 (17.3) 0.27 Renal failure (Creatinine ≥2.2 mg/dl, %) 21 (7.6) 9 (11.1) 12 (6.1) 0.21 Major neurological dysfunction 36 (12.9) 9 (11.1) 27 (13.7) 0.70 Coronary artery disease 173 (62.2) 57 (70.4) 116 (58.9) 0.08 Acute MI (<90days, %) 38 (13.7) 14 (17.3) 24 (12.2) 0.26 Critical perioperative state 1 (0.4) 0 1 (0.5) 1.0 Previous CABG 45 (16.2) 11 (13.6) 34 (17.3) 0.59 Previous Valve Surgery 10 (3.6) 3 (3.7) 7 (3.6) 1.0 Mitral valve lesion or defect (≥II°, %) 38 (13.7) 14 (17.3) 24 (12.2) 0.26 Ejection Fraction <40% 31 (11.2) 14 (17.3) 17 (8.6) 0.06 Euro-SCORE in% 22.35±15.55 21.90±12.64 22.54±16.63 0.76 Valves Edwards SAPIEN 203 (73) 64 (79) 139 (70.6) 0.18 Medtronic CoreValve 75 (27) 17 (21) 58 (29.5) Legend: MI, myocardial infarction; CABG, coronary artery bypass graft; COPD, chronic obstructive pulmonary disease

27