Author + information
- Received January 11, 2018
- Revision received June 8, 2018
- Accepted June 12, 2018
- Published online August 6, 2018.
- Antonio Mangieri, MDa,
- Giuseppe Lanzillo, MDa,
- Letizia Bertoldi, MDa,
- Richard J. Jabbour, MDb,
- Damiano Regazzoli, MDa,
- Marco B. Ancona, MDa,
- Akihito Tanaka, MDa,
- Satoru Mitomo, MDa,
- Stefano Garducci, MDc,
- Claudio Montalto, MDa,
- Matteo Pagnesi, MDa,
- Francesco Giannini, MDa,
- Manuela Giglio, MDd,
- Matteo Montorfano, MDa,
- Alaide Chieffo, MDa,
- Josep Rodès-Cabau, MDe,
- Fabrizio Monaco, MDa,
- Gabriele Paglino, MDa,
- Paolo Della Bella, MDa,
- Antonio Colombo, MDa and
- Azeem Latib, MDa,f,∗ ()
- aIRCCS San Raffaele Scientific Institute, Milan, Italy
- bImperial College, London, United Kingdom
- cAzienda Socio-Sanitaria Territoriale, Vimercate, Italy
- dIstituto Clinico Sant’Ambrogio, Milan, Italy
- eQuebec Heart & Lung Institute, Laval University, Quebec City, Quebec, Canada
- fDivision of Cardiology, Department of Medicine, University of Cape Town, Cape Town, South Africa
- ↵∗Address for correspondence:
Dr. Azeem Latib, San Raffaele Scientific Institute, Via Olgettina 60, 20132 Milan, Italy.
Objectives This study sought to determine predictors of advanced conduction disturbances requiring late (≥48 h) permanent pacemaker replacement (PPM) after transcatheter aortic valve replacement (TAVR).
Methods Data of consecutive patients were identified by retrospective review of a TAVR database of a single center in Milan, Italy, between October 2007 and July 2015. We defined delta PR (ΔPR) and delta QRS (ΔQRS) interval as the difference between the last PR and QRS length available 48 h after TAVR and the baseline PR and QRS length.
Results Overall population included 740 patients. We excluded 78 patients who already had a PPM and 51 patients who received a PPM <48 h after TAVR. The final analysis included 611 patients. Fifty-four patients (8.8%) developed an advanced conduction disturbance requiring PPM ≥48 h following TAVR. Patients who required a late PPM implant had a wider QRS width (113 ± 25 ms vs. 105 ± 23 ms; p = 0.009) and a higher prevalence of baseline right bundle branch block (12.9% vs. 5.3%; p = 0.026) and were more likely to have a self-expandable valve implanted (51.8% vs. 31.9%; p = 0.003). The ΔPR was 40 ± 51 ms (p = 0.0001) and the ΔQRS was 22 ± 61 ms (p = 0.001). Multivariable analysis revealed that baseline right bundle branch block (odds ratio: 3.56; 95% confidence interval: 1.07 to 11.77; p = 0.037) and ΔPR (odds ratio for each 10-ms increase: 1.31; 95% confidence interval: 1.18 to 1.45; p = 0.0001) are independent predictors of delayed advanced conduction disturbances.
Conclusions This analysis showed that baseline right bundle branch block and the amount of increase of PR length after TAVR are independent predictors of late (≥48 h) advanced conduction disturbances requiring PPM replacement after TAVR in this cohort. A simple ECG analysis could help in detecting potentially lethal advanced conduction disturbances that could occur more than 48 h after TAVR.
- aortic stenosis
- left bundle branch block
- permanent pacemaker
- right bundle branch block
- transcatheter aortic valve replacement
Transcatheter aortic valve replacement (TAVR) has evolved from being a novel technology to a robust therapy for patients experiencing from severe aortic stenosis (1). Since the first implant performed in 2002, TAVR has evolved into a mature therapy with significant improvement in procedural and mid-term outcomes and a compelling alternative to surgery in the intermediate-risk population. Its widespread use is still limited by a few clinically relevant issues (2), such as conduction disturbances. However, with increasing operator experience and technological improvements, the rate of complications is reducing, and TAVR has become a more predictable and standardized procedure, resulting in patients being discharged earlier with positive effects in terms of post-procedural length of stay and related costs. The post-procedural length of stay is one of the main cost components in the early period after TAVR and significantly influenced by the necessity of prolonged monitoring of patients who developed a new conduction disturbance (3). Patients developing such electrocardiographic (ECG) alterations are at higher risk of sudden cardiac death and often require a period of monitoring to ascertain whether a permanent pacemaker replacement (PPM) is needed (4). As a consequence, hospitalization may be prolonged and the recovery of these patients is hampered. Several anatomic, procedural, and electrophysiological factors have been identified as predictors of the need of PPM after TAVR (5). However, limited data are available about the predictors of late conduction disturbances following TAVR. Accordingly, we sought to find clinical and ECG predictors of late advanced conduction disturbances (≥48 h) requiring PPM after TAVR that can be easily used in daily routine clinical practice.
All consecutive patients who underwent TAVR between January 2009 and July 2015 were identified by a retrospective review of medical records and dedicated TAVR database at a high-volume center in Milan, Italy. In the final analysis, we excluded patients who already had a PPM, and patients who received a PPM <48 h after TAVR.
The decision to perform TAVR was based on the severity of symptoms, risk evaluation, and contraindications to surgery. All patients were evaluated by a multidisciplinary heart team comprised of cardiologists, interventional cardiologists, cardiothoracic surgeons, and cardiac anesthetists. Patient demographics, symptoms, and comorbidities were documented, and individual risk was calculated by the logistic European System for Cardiac Operative Risk Evaluation (EuroSCORE) and Society of Thoracic Surgeons predicted risk of mortality score. Transthoracic echocardiography was the initial screening examination used to evaluate severity of aortic stenosis. All patients with severe aortic stenosis considered for an intervention underwent computed tomography to evaluate annular size, coronary arteries, and peripheral access sites. Coronary angiography was selectively performed in patients with abnormal coronaries or suboptimal computed tomography visualization. The standard access route was the transfemoral approach.
ECG records were collected at baseline, immediately after TAVR, and in the following days. Post-procedural ECGs were analyzed by 4 cardiologists (A.M., G.L., L.B., and M.P.) blinded to whether patients received late PPM replacement. Post-procedural ECG data were obtained analyzing the last ECG recorded within the first 48 h. We defined delta PR (ΔPR) and delta QRS (ΔQRS) interval as the difference between the last PR and QRS length available 48 h after TAVR and the baseline PR and QRS length. The diagnosis of conduction abnormalities was based on the task force on cardiac pacing and resynchronization therapy of the European Society of Cardiology developed in collaboration with the European Heart Rhythm Association (6). We defined patients with delayed advanced conduction disturbances as those who first developed advanced heart block 48 h or later TAVR. Patients with new-onset conduction disturbances were reviewed by an electrophysiologist (G.P.) and PPM replacement indicated in presence of advanced conduction delays that were expected not to resolve according to the current guidelines recommendations (6). The selection of a single- or dual-chamber pacemaker was left to the discretion of the implanting physician. None of the patients were discharged within the first 48-h after TAVR. The general policy regarding discharge after TAVR at our institution is the following:
• Early discharge (48 to 72 h): feasible in presence of normal EKG or left bundle branch block (LBBB) without PR prolongation, baseline good clinical features (no chronic kidney disease, normal or mildly depressed ejection fraction, uncomplicated TAVR procedure).
• Late discharge (up to 72 h): significant changes in the atrioventricular (AV) and intraventricular conduction, acute kidney injury, patients with compromised ejection fraction, complicated TAVR procedure.
All patients were scheduled for outpatient transthoracic echocardiography to assess aortic prosthesis function and to check for any evidence of intravalvular or paravalvular leaks. Follow-up data were obtained either by telephonic interview or outpatient visit. Each patient provided written informed consent for the procedure and data collection.
Continuous variables are presented as mean ± SD and were compared with the unpaired Student’s t-test or Mann-Whitney U test. Categorical variables are presented as numbers and percentages and were analyzed using chi-square test and Fisher exact test. Paired Student's t-test was used to compare pre- and post-procedural PR and QRS lengths. Multivariable logistic regression analysis was performed to identify independent predictors of delayed PPM replacement. All variables with a univariable p value < 0.10 or if they had clinical relevance were subsequently entered into the final model. Results of the logistic regression analysis are presented as odds ratio with 95% confidence interval. The accuracy of logistic regression model was tested using C-statistics. Hosmer-Lemeshow goodness-of-fit tests were performed to assess the fit of the model. Kaplan-Meier analysis was used for follow-up data. The sample size was calculated on the basis of the hospitalization rate at 1-year follow-up. The standard sample size calculation for the 2-sample Student's t-test was performed for the transformed variable. The following assumptions were taken into account: at 1-year after TAVR, the median hospitalization rate is 32% among patients with PPM and 27% without PPM. These assumptions were based on previous studies evaluating the follow-up of patients after TAVR. A sample size of 441 patients was needed to reject the null hypothesis with power of 90% (β = 0.90) and a 2-sided α value of 0.05.
All reported p values were 2-sided and a p < 0.05 was considered statistically significant. Statistical analyses were performed using SPSS version 16.0.2 (SPSS Inc., Chicago, Illinois) and GraphPad Prism software version 4 (GraphPad, Inc., San Diego, California).
The overall population included 740 patients. We excluded 78 patients who already had a PPM and 51 patients who received a PPM <48 h after TAVR. The final analysis included 611 patients with 54 patients (8.8%) requiring a PPM ≥48 h after TAVR for advanced conduction disturbances (Figure 1). PPM replacement was performed after a mean time of 6.1 ± 3.9 days after TAVR (Figure 2). Reasons for PPM replacement included: 1) complete AV block (n = 42; 77%); 2) Mobitz type II block (n = 4; 7%); 3) pathological pauses and asystole (n = 7; 6.6%); or 4) a symptomatic junctional rhythm (n = 1; 1.8%).
The baseline characteristics of the included cohort are reported in Table 1. There were no significant differences in clinical characteristics between patients who did or did not receive a late PPM.
Patients who developed advanced conduction disturbances after TAVR had a wider QRS width at baseline (113 ± 25 ms vs. 105 ± 23 ms; p = 0.009). Baseline right bundle branch block (RBBB) was also more frequent in patients receiving a late PPM for advanced AV block (35.1 ± 19.7%; p = 0.026). After the development of an advanced AV block the mean time for PPM implant was 9 ± 7 h. The median time of ECG collection was 41 h (interquartile range: 33.7 to 46.3 h). Patients who required a late PPM were more likely to have self-expandable valves implanted (35.1% vs. 53.3%; p = 0.011) and to undergo post-dilatation (31.4% vs. 17.9%; p = 0.017). Other pre-procedural ECG features and procedural features are reported in Table 2. PR length and QRS width after TAVR were significantly larger among patients who developed late conduction disturbances requiring PPM (p = 0.0001 for PR length; p = 0.005 for QRS width). Comparing the baseline and post-procedural ECGs of patients who received a late PPM, QRS and PR interval widened significantly with a ΔPR of 40 ± 51 ms (p = 0.0001) and a ΔQRS of 22 ± 61 ms (p = 0.001) (Figure 3).
Multivariable analysis (Table 3) revealed that baseline RBBB (odds ratio: 3.54; 95% confidence interval: 1.07 to 11.77; p = 0.037) and ΔPR (odds ratio for each 10-ms increase: 1.31; 95% confidence interval: 1.18 to 1.45; p = 0.0001) were independent predictors of late PPM replacement. The Hosmer-Lemeshow goodness-of-fit test was not statistically significant (p = 0.67), suggesting adequate model fit. The predictive value of the multivariate analysis for late PPM implant was confirmed by C-statistics (0.768).
At 1-year follow-up, no differences in terms of all-cause mortality were found between the 2 groups; however, patients who received a late PPM had a higher rate of rehospitalization for cardiovascular causes (hazard ratio: 4.01; 95% confidence interval: 2.5 to 50.1; p = 0.0015) (Figure 4).
The main findings of this study are: 1) the development of advanced conduction disturbances ≥48 h after TAVR requiring PPM replacement is frequent, occurring in 8.8% of patients; 2) patients who developed late advanced AV block were more likely to have a significant widening of the PR and QRS intervals (40 ± 51 ms and 22 ± 61 ms, respectively) after TAVR; 3) at multivariable analysis, ΔPR and baseline RBBB after TAVR are predictors of delayed advanced conduction disturbance; and 4) patients that had a late PPM experienced a higher rate of cardiovascular rehospitalization when compared with those that did not at 1-year follow-up.
Our report has potentially important clinical implications and suggests that the simple analysis of the ECG could help to recognize patients at higher risk of advanced AV block after TAVR, thus preventing potentially life-threatening complications and identifying patients that would benefit from a PPM.
TAVR is now a proven alternative to conventional surgery in intermediate- and high-risk patients with data starting to appear in low-risk patients (7). Several intraprocedural and post-procedural complications remain and are of concern when moving to low-risk patients. One of these is the occurrence of advanced conduction delays that, if left untreated, can be responsible for sudden cardiac death after discharge. Landmark studies have clarified that patients with pre-existing first-degree AV block, RBBB, and left hemi-block have an increased risk of PPM replacement after TAVR (8). However, these studies considered periprocedural PPM implant, which occurs immediately after the replacement of the valve, and late PPM implant, which is required days after TAVR. Moreover, few studies evaluate the dynamic ECG changes after the procedure. What is known is that in the absence of any new AV and new intraventricular conduction delays, the risk of post-procedural PPM replacement seems to be low (9). However, in the case of new conduction delays after TAVR, no data are available about the risk of late PPM replacement. For the first time, we established that an important predictor of delayed PPM replacement is the PR length after the procedure. PR widening has been described after CoreValve but has not been associated with PPM replacement (10). Conversely, we found that prolongation of PR interval following TAVR is a predictor of advanced AV block requiring PPM. Notably, observational studies in the Framingham cohort showed that healthy people with a prolonged AV interval have a 3-fold increase of PPM at follow-up irrespective of age and QRS length (11). Prolonged PR length after TAVR could be as a result of pre-existing fibrosis or calcification of the cardiac skeleton, which could trigger advanced AV block following valve replacement. Similarly, we found that also baseline RBBB after TAVR is a significant predictor of delayed advanced AV block.
A recent study by Watanabe et al. has demonstrated the prognostic importance of baseline RBBB as it was correlated with an increased risk of sudden death and cardiovascular mortality after TAVR in patients treated with a balloon-expandable valve (12). Our study validates this finding and highlights that patients with baseline RBBB should not be considered for early discharge and require longer monitoring to detect delayed AV conduction disturbances. In our population, persistent new-onset LBBB after TAVR was found in 19% of patients, which is consistent with previous reports (13,14). However, new-onset LBBB was not associated with an increased risk of PPM replacement after multivariable analysis. Several studies have found a correlation between new-onset LBBB and the risk of PPM replacement (15). However, these studies failed to report in detail the PPM strategy and timing. This is important, because it is possible that complete AV block resolves after TAVR and complete LBBB may remain. If a PPM is implanted early, those patients would not show up in the group of patients with new-onset LBBB at hospital discharge, because they were excluded in our study (16).
At 1-year follow-up (median follow-up time, 365 days; interquartile range: 356 to 365 days), the delayed PPM group had an increased risk of cardiovascular hospitalization compared with patients who did not received a PPM, confirming a previous report in which the chronic presence of PPM in TAVR population was associated with a higher rate of hospitalization (17). One hypothesis is that the chronic stimulation induced by the PPM could have a detrimental effect on left ventricular function, especially in patients with structural heart disease, thus increasing the risk of heart failure and poor recovery of functional status (5).
Delayed conduction disturbances: Risk and pitfalls of the early discharge strategy
Our study clearly demonstrates that the risk of delayed advanced conduction disturbances after TAVR is not negligible (8.8%) and points out the possible risks related to an early discharge strategy. Delayed conduction disturbances can be potentially fatal because it manifests in most cases as complete AV block. As previously demonstrated by Urena et al. (3), patients developing prolonged LBBB are at higher risk of sudden death after discharge. For this reason, an early discharge strategy should be cautiously evaluated and reserved only for low-risk patients with evidence of normal baseline and post-procedural ECG.
First, although the ECGs were evaluated by experienced cardiologists blinded to whether a late PPM was implanted, there was no centralized core laboratory available for analysis. Second, this was a single-center retrospective study with a relatively low number of patients. Because of the retrospective design of the study, a predefined protocol of ECG acquisition was not available.
This preliminary analysis suggests that the development of late (≥48 h) advanced conduction disturbance requiring a PPM after TAVR is relatively frequent (8.8% of patients) and is associated with a higher rate of cardiovascular rehospitalization. The amount of increase of PR length and baseline RBBB are independent predictors of late PM replacement after TAVR. Careful monitoring to detect fatal arrhythmic events after TAVR should be conducted especially in patients with evidence of PR length prolongation after TAVR and in those with baseline evidence of RBBB. Larger studies are needed to confirm these findings.
WHAT IS KNOWN? Pacemaker replacement is a frequent complication following TAVR with limited data regarding delayed conduction disturbances.
WHAT IS NEW? In our single-center experience, we found that the prevalence of conduction disturbances requiring a pacemaker replacement 48 h after TAVR is 8.8% and that baseline RBBB and prolongation of PR interval are independent predictors of delayed conduction disorders.
WHAT IS NEXT? To establish the optimal period of observation for those patients at higher risk of delayed conduction disturbances.
Dr. Rodès-Cabau has received institutional research grants from Edwards Lifesciences and Medtronic. Dr. Latib is on the Medtronic Advisory Board; and has received speaking honoraria from Abbott. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- left bundle branch block
- permanent pacemaker replacement
- right bundle branch block
- transcatheter aortic valve replacement
- Received January 11, 2018.
- Revision received June 8, 2018.
- Accepted June 12, 2018.
- 2018 American College of Cardiology Foundation
- Urena M.,
- Webb J.G.,
- Tamburino C.,
- et al.
- Barbanti M.,
- Capranzano P.,
- Ohno Y.,
- et al.
- Nazif T.M.,
- Dizon J.M.,
- Hahn R.T.,
- et al.
- Brignole M.,
- Auricchio A.,
- Baron-Esquivias G.,
- et al.
- Jabbour R.J.,
- Pagnesi M.,
- Kawamoto H.,
- et al.
- Siontis G.C.,
- Jüni P.,
- Pilgrim T.,
- et al.
- Toggweiler S.,
- Stortecky S.,
- Holy E.,
- et al.
- Watanabe Y.,
- Kozuma K.,
- Hioki H.
- Urena M.,
- Mok M.,
- Serra V.,
- et al.
- Houthuizen P.,
- Van Garsse L.A.,
- Poels T.T.,
- et al.
- Regueiro A.,
- Abdul-Jawad Altisent O.,
- Del Trigo M.,
- et al.
- Roten L.,
- Meier B.
- Dizon J.M.,
- Nazif T.M.,
- Hess P.L.,
- et al.