Impact of New-Onset Conduction Disturbances following Transcatheter Aortic Valve Replacement on Outcomes: A Single-Center Study

Background Transcatheter aortic valve replacement (TAVR) is known to increase the incidence of conduction disturbances compared to surgical aortic valve replacement; however, there are limited data on the impact and duration of these conduction disturbances on longer term outcomes. Objective To determine the differential impact of persistent versus nonpersistent new-onset conduction disturbances on TAVR-related complications and outcomes. Methods This is a single-center retrospective analysis of 927 consecutive patients with aortic stenosis who underwent TAVR at Yale New Haven Hospital from July 2012 to August 2019. Patients with new-onset conduction disturbances within 7 days following TAVR were selected for this study. Persistent and nonpersistent disturbances were, respectively, defined as persisting or not persisting on all patient ECGs for up to 1.5 years after TAVR or until death. Results Within 7 days after TAVR, conduction disturbances occurred in 42.3% (392/927) of the patients. Conduction disturbances persisted in 150 (38%) patients and did not persist in 187 (48%) patients, and 55 (14%) patients were excluded for having mixed (both persistent and nonpersistent) disturbances. Compared with nonpersistent disturbances, patients with persistent disturbances were more likely to receive a PPM within 7 days after the TAVR procedure (46.0% versus 4.3%, p < 0.001) and had a greater unadjusted 1-year cardiac-related and all-cause mortality risk (HR 2.54, p=0.044 and HR 1.90, p=0.046, respectively). Conclusion Persistent conduction disturbances were associated with a greater cardiac and all-cause mortality rate at one year following TAVR. Future research should investigate periprocedural factors to reduce persistent conduction disturbances and outcomes beyond one year follow-up.


Introduction
Transcatheter aortic valve replacement (TAVR) is approved as an alternative to surgical aortic valve replacement across the spectrum of surgical risk [1,2]. Despite advances in valve technology and implantation techniques, conduction disturbances are a known complication of TAVR related to clinical and procedural factors, including implantation depth, valve oversizing, and valve type [3]. Conduction disturbances are reported to occur in 31-45% of the patients depending on the type of valve implanted [4][5][6], with newonset left bundle branch block (LBBB) reported in 7-65% [7,8] and new-onset atrial fbrillation in 5-13% of patients [1,2,[9][10][11], and resulting in permanent pacemaker (PPM) implantation in cases of high degree atrioventricular block (AVB) [12,13]. Prior studies have shown that persistent new-onset conduction disturbances after TAVR are associated with worse outcomes, including increased risk of cardiac and all-cause mortality with new-onset LBBB [13,14] and increased risk of all-cause mortality with newonset atrial fbrillation [15]. However, the diferential prognostic impact of persistent versus nonpersistent newonset disturbances on TAVR-related complications and outcomes is not well understood. We address this question in a single-center TAVR registry.

Methods
Tis is a single-center, retrospective analysis of consecutive patients with severe symptomatic aortic stenosis who underwent TAVR at Yale New Haven Hospital from July 2012 to August 2019. We include only patients who had newonset conduction disturbances within 7 days after TAVR. Patients were excluded if they had a pre-existing PPM and/ or implantable cardioverter defbrillator (ICD) or if electrocardiogram (ECG) was not performed or available before or within 7 days after TAVR. Patients with new-onset disturbances after 7 days after TAVR with no new-onset disturbances or with mixed (both persistent and nonpersistent) disturbances were also excluded ( Figure 1).
For all patients, ECGs were performed before, immediately after, and up to 1.5 years after TAVR. Conduction defect is defned on ECG as left or right bundle branch block, intraventricular conduction delay, 2 : 1, 3 : 1, and 4 : 1 block, bifascicular block, left anterior and posterior fascicular block, frst-degree atrioventricular block, second degree atrioventricular block Mobitz Type I and II, third-degree atrioventricular block, and atrial fbrillation or futter. Newonset persistent disturbances were defned as new disturbances that persisted on all ECGs for up to 1.5 years after TAVR or until death. New-onset nonpersistent disturbances were defned as disturbances that did not persist on all ECGs for up to 1.5 years after TAVR or until death. All ECG readings were conducted by independent board-certifed cardiologists and entered into the electronic medical record (EMR). Chart review was conducted by querying the EMR database. Te fndings were classifed based on the American College of Cardiology/American Heart Association/Heart Rhythm Society recommendations [16]. PPM placement was clinically based by the TAVR operator in consultation with an electrophysiologist.
Data acquisition, monitoring, and outcome assessments were performed according to the STS/ACC Transcatheter Valve Terapy (TVT) Registry [17,18]. Eligibility for TAVR was based on decision of the multidisciplinary heart team consisting of experienced surgeons, interventional cardiologists, and imaging specialists. Tis study was approved by the Yale Institutional Review Board (No. 2000028604). Device success was defned as successful vascular access, delivery, and deployment of a single device in the proper anatomic location, appropriate performance of the prosthetic heart valve (aortic valve area >1.2 cm 2 and mean aortic valve gradient <20 mm Hg or peak velocity <3 m/s without moderate or severe prosthetic valve aortic regurgitation) and the successful retrieval of the delivery system. Implant success was defned as the correct positioning of a single device in the proper anatomic location.
Te primary endpoint was all-cause mortality at 1 year after TAVR. Secondary endpoints included the major adverse cardiac events (MACEs) at 30 days, defned as a composite of death, myocardial infarction (MI), stroke, valve-related hospitalization, and cardiac arrest, according to the Valve Academic Research Consortium-2 (VARC-2) defnitions [19]. Other adverse events at 30 days included major bleeding, major vascular complications, and all-cause mortality. Major vascular complication was defned as a composite of the major vascular access site complication or unplanned vascular surgery, annular rupture, aortic dissection, or perforation with or without tamponade. All 30day adverse events were adjudicated by board-certifed cardiologists using a combination of site-reported clinical information and targeted chart reviews. Mortality and cause of death at 1 year were determined by the chart review of electronic medical records, publicly accessible online obituaries, and the Centers for Disease Control and Prevention National Death Index registry.

Statistical Analysis.
Descriptive statistics for continuous variables included mean, standard deviation, and sample size for each treatment group. Categorical variables were summarized using frequencies, percentages, and sample size for each treatment group. Categorical variables were compared using the Pearson χ 2 test or the Fisher exact test. Continuous variables were compared with Student's t-test or the Wilcoxon rank-sum test if the data failed to meet the assumption for normality per the Shapiro-Wilk test. For time-to-event data, Kaplan-Meier estimates were calculated and displayed graphically. Te univariable Cox proportional hazard regression model presented hazard ratios (HRs) with 95% confdence intervals. All analyses were conducted using SAS version 9.4 (SAS Institute). No imputation was considered for missing values. Values of p < 0.05 were considered statistically signifcant.

TAVR Procedural Characteristics and Outcomes.
Procedural TAVR characteristics of the study population are summarized in Table 1. Most cases were elective (89.9%) and performed via transfemoral access (89.9%) under moderate sedation (61.9%) or general anesthesia (36.9%). Overall device success was 97.9% and implantation success was 98.2%, with no diferences in the two groups. Te use of antiplatelet and antithrombotic medications at discharge was similar between the two groups.

Outcomes at 30
Days. Te occurrence of MACE at 30 days post procedure was low and similar in both groups ( Table 2). Tere were no signifcant diferences between the persistent and nonpersistent groups for 30-day unadjusted or adjusted rates of MACE and mortality (    (Figure 2).

PPM Implantation.
Patients with persistent disturbances were more likely to receive a PPM within 7 days after the TAVR procedure compared with patients with nonpersistent disturbances (46.0% versus 4.3%, p < 0.001) ( Figure 3). Patients with persistent disturbances had greater 30-day PPM rates after TAVR compared with the patients with nonpersistent disturbances (46.7% versus 5.9%, p < 0.001). In the persistent group, the indications for 30-day PPM placement were complete AVB, bundle branch block, and primary prevention in 90%, 8%, and 2% of the patients, respectively. In the nonpersistent group, the indications for 30-day PPM were complete AVB, high-grade AVB, and bundle branch block in 70%, 20%, and 10% of the patients, respectively. One-year PPM rates according to the valve type were 12.8% for balloon-expandable valves and 11.0% for self-expanding valves.

. Discussion
Tis study compared characteristics and outcomes for patients undergoing TAVR who had persistent versus nonpersistent new-onset conduction disturbances postprocedure. Te main fndings of our study are as follows: (1) 38% of the patients with new-onset conduction disturbances after TAVR had persistent disturbances and 48% had nonpersistent disturbances, (2) persistent disturbance patients were more likely to receive a PPM within 7 days after the TAVR procedure, and (3) persistent disturbance patients had a greater unadjusted 1-year mortality risk.
Prior studies have shown that persistent new-onset conduction disturbances after TAVR are associated with worse outcomes, including the increased risk of cardiac mortality and PPM implantation at 1-year follow-up for new-onset LBBB [14], increased all-cause and cardiovascular mortality at 2-year follow-up for new-onset persistent LBBB [13], and increased mortality at a median follow-up of 305 days for new-onset atrial fbrillation [15]. However, this is the frst study to compare the diferential impact of persistence on outcomes.
In our study, 38% of the patients with new-onset disturbances had persistent only disturbances, which is similar to prior smaller scale reports with the SAPIEN valve [20]. Persistent conduction disturbances were associated with an increased risk for 1-year cardiac-related and all-cause mortality based on univariable analysis (Table 2; Figure 2 Kaplan-Meier). Additionally, persistent disturbance patients had signs of worse baseline health; greater 5-meter walk score, more prior conduction disturbances, and lower baseline LVEF (Table 1).
Our study includes newer generation TAVR systems (Edward Sapien 3 valve and Medtronic Evolut R and Pro valves), which should have improved valve design and implantation techniques, lower pacemaker rates [21], and  Journal of Interventional Cardiology better outcomes [22][23][24][25]. Our study also includes intermediate STS risk patients, an indication expanded in 2016 after the PARTNER-2 SURTAVI trial [26]. Te results of our study suggest that the persistence of disturbances may outweigh these benefcial changes and contribute to increased mortality. Additionally, the similar STS mortality risk scores between persistent and nonpersistent disturbance patients (6.53% vs. 6.30%, respectively, p � 0.62) suggest that the persistent nature of the disturbances may play a bigger role in mortality than expected.
Te persistence of conduction disturbances may be explained by modifable causes such as mechanical trauma from the implantation depth, heart valve oversizing, and radial force of the heart valve at the left ventricular outfow tract (LVOT) level or nonmodifable causes such as LVOT geometry, anatomical variability of the conduction system, and distribution and amount of calcifcation [27]. Further research is needed to assess and educate operators on the factors contributing to persistent conduction disturbances.

Limitations
Our study represents a real-world population of patients undergoing TAVR at a single-center tertiary referral center. As a retrospective study, it has inherent limitations and is subject to bias, and being conducted at a single center, the results may not be generalizable. Nonpersistent conduction disturbances that arose immediately after TAVR but resolved before the postprocedure ECG were not captured, and there are likely other unmeasured confounders afecting 30-day and 1-year survival. Comparison of individual conduction disturbances and multivariable regression analysis could not be conducted because the respective numbers of patients and outcome events were too low. Longer follow-up past 1 year or a larger patient population may be needed to improve statistical power. Lastly, we did not examine diferences in outcomes according to newer versus older generation TAVRs.

Conclusion
Tis single-center study suggests that persistent conduction disturbances after TAVR lead to worse 1-year outcomes when compared with nonpersistent disturbances. Additionally, these results suggest that patients with new-onset persistence disturbances within seven days after TAVR have greater PPM/ICD rates when compared with patients with nonpersistent new disturbances after TAVR. Future research should look at periprocedural factors to reduce persistent conduction disturbances in TAVR patients.

Data Availability
Te participants of this study did not give written consent for their data to be shared publicly, so due to the sensitive nature of the research, supporting data are not available.

Conflicts of Interest
Te authors declare that they have no conficts of interest. Dr. Ryan Kaple is on the speaker bureau for Abbott and Edwards Lifesciences. Dr. John Forrest reported grants and personal fees from Medtronic Inc and Edwards Lifesciences outside the submitted work.