Abstract: Background. The need for permanent pacemaker implantation (PPI) following transcatheter aortic valve implantation (TAVI) is a recognized complication due to compression of the cardiac conduction system. PPI rates vary according to type of TAVI device, implantation technique, and patient-related factors. Methods. We analyzed the baseline characteristics, and then the preprocedure and daily postprocedure electrocardiograms (ECGs) of 101 consecutive patients who underwent TAVI using the Lotus valve (Boston Scientific) between 2013 and 2015. Results. Mean patient age was 81.7 ± 7.7 years, 56% were male. and 21 patients were excluded because of baseline atrial fibrillation or a pre-existing PPI. Of the remaining 80 patients, a total of 28 (35%) underwent PPI at a mean 2.7 days after TAVI. There were no differences in development of new left bundle-branch block between the two groups (65% PPI group vs 74% no-PPI group; P=.37). The initial PR interval did not differ between groups (183 ms PPI group vs 184 ms no-PPI group). The PR interval increased by day 1 post procedure (232 ms PPI group vs 195 ms no-PPI group; P<.01) and day 2 (267 ms PPI group vs 211 ms no-PPI group; P<.05). Conclusions. Patients with a PR interval ≤230 ms on day 1 or ≤260 ms on day 2 post TAVI using the Lotus device are at very low risk of requiring PPI, irrespective of the presence of left bundle-branch block, and can be safely discharged without prolonged monitoring.
J INVASIVE CARDIOL 2020 July 10 (Epub Ahead of Print).
Key words: ECG, pacemaker, TAVR
Transcatheter aortic valve implantation (TAVI) is now a well-established procedure.1 Disturbance of the conduction system during and immediately after TAVI is common, due to compression of the intrinsic cardiac conduction tissue by the device. We examined the electrocardiograms (ECGs) of patients before and after TAVI using the Lotus “classic” device (Boston Scientific) to help predict those who would require permanent pacemaker implantation (PPI) in the postprocedure phase.
The atrioventricular node and His-Purkinje conduction fibers run close to the aortic annulus.2 Calcification of the aortic valve and annulus can extend into this region and has been associated with an increased incidence of conduction delay following surgical aortic valve replacement (SAVR).3,4 Pacemaker rates following SAVR average ~5%,3 but rates of transient heart block are much higher, at up to 17%.5
Risk factors for PPI following SAVR include female sex, hypertension, impaired preoperative left ventricular systolic function, annular calcification, bicuspid anatomy, preprocedure bundle-branch block, prolonged aortic cross-clamp and perfusion times, preoperative use of calcium-channel blockers, and continuous suture techniques.3,5,6
Risk factors in the TAVI cohort have features in common, but height of implantation and device type are also important factors.7 PPI rates using the Sapien XT valve (Edwards Lifesciences) are as low as 5% if high implantation is achieved,8 but have been reported in up to 19%.9 In one large registry, PPI rate was 17% using the CoreValve (Medtronic)7 and 32% using the Lotus valve.10
The need for PPI often happens late after valve implantation. The ability to predict those who would require PPI could potentially save bed days and costs, allowing earlier safe discharge for those who are unlikely to require pacing.
We analyzed 101 consecutive patients undergoing a TAVI procedure using the Lotus device between September 2013 and May 2015. Baseline demographics were recorded, along with analysis of the 12-lead ECG immediately prior to the procedure, post procedure, and daily until discharge.
Patients with atrial fibrillation were excluded from the analysis of PR intervals, as this is not possible to calculate. Patients with pre-existing pacemakers were also excluded from the analysis. The decision to perform PPI was at the treating physician’s discretion.
Mean patient age was 81.7 ± 7.7 years and 56% were men. One patient had a balloon valvuloplasty alone, and 2 patients were lost to follow-up. There were no deaths. Of the remaining 98 patients, 18 had a pre-existing PPI prior to the TAVI procedure and thus were discounted from the subsequent analysis. In 3 of these patients, preprocedural PPI was elective due to existing right bundle-branch block.
Of the 80 patients remaining for analysis, 28 (35%) went on to have PPI during their admission for TAVI, on average 2.7 days post TAVI (range, 0-8 days). No patients who were in sinus rhythm before the procedure developed atrial fibrillation. Nine patients (11.3%) developed procedural complete heart block; 8 of these had subsequent PPI, 1 of whom developed restoration of sinus rhythm. Among the remaining 71 patients, preprocedural PR interval was similar between the two groups (Table 1). The immediate postprocedure ECGs of those who went on to have PPI had a PR interval that was not statistically elongated (200 ms in the PPI group vs 214 ms in the no-PPI group; P=.15).
Indications for PPI in the post-TAVI period were: complete atrioventricular block (n = 17), prolonged first-degree atrioventricular block with left bundle-branch block (n = 9), and atrial fibrillation with slow ventricular response (n = 2).
The PR interval in the no-PPI group stayed similar on days 1 and 2 post procedure (195 ms and 211 ms, respectively). The average PR interval for this cohort then extended on day 3 to 233 ms; however, there is a large selection bias at this stage, as half of this group (26 of 52) was discharged on day 2.
In the PPI group, the PR interval on day 1 was significantly longer than in the no-PPI group (232 ms vs 195 ms, respectively; P<.01). The PR interval increased on day 2 to 267 ms and on day 3 to 283 ms, and remained at 283 ms on day 4. However, again there is significant bias by this stage, as many patients underwent semielective PPI (at the treating physician’s discretion) usually once they had left bundle-branch block and a PR interval of 280 ms.
There were no differences in QRS duration between the two groups before or after TAVI (106 ms in the PPI group vs 103 ms in the no-PPI group [P=.59] and 145 ms in the PPI group vs 142 ms in the no-PPI group [P=.64]), respectively, and no difference in the incidence of new left bundle-branch block (65% in the PPI group vs 74% in the no-PPI group; P=.37).
As TAVI becomes more accepted and moves into the cohorts of patients at low or moderate risk, the requirement for PPI will be one of the ever-decreasing number of factors that will influence both length of stay and complication rates from the procedure. In this cohort of high surgical risk patients, it seems that almost all patients have some disturbance of the conduction system. In most patients, this is transient and not clinically significant, but in a significant proportion it has implications on management and future care.
TAVI, as a procedure, has now matured to such a degree in many experienced centers that once the vascular access site is secure patients are often suitable for discharge on day 1 or 2 post procedure. In this scenario, and when TAVI is opened up to a lower risk cohort, having to remain in hospital for observation of the conduction system will have major implications in terms of hospital length of stay, resource utilization, and health economics. The ability to discharge with confidence those patients at very low risk of PPI will be a significant advantage.
This also has implications for the use of temporary pacing wires, which are routine in many centers and left in situ for a variable length of time. The ability to avoid using a temporary pacing wire at all, or removing it very early, would have several advantages including improved early mobilization and recovery of patients, as well as reduced infection risk.
From this study, it may be reasonable to assume that any patient whose PR interval is <230 ms on day-1 ECG is in a very low-risk group and could be discharged safely. Those with PR interval between 230-260 ms may need observation until day 3 post procedure, and it would be reasonable to discharge at that time if the PR interval remains stable. Those with PR interval >260 ms are in the high-risk group and should be monitored closely. The majority of patients had interventricular delay (left bundle branch block); however, this did not appear to impact upon need for PPI.
The need for PPI is more difficult to predict in patients with atrial fibrillation. Given that patients in sinus rhythm usually need 3 days to “declare” themselves as needing a PPI, it may be reasonable to extrapolate this factor to those with atrial fibrillation such that if they are stable on their usual medication doses on day 3, they are at low risk for needing PPI. However, further investigation into PPI rates in those with atrial fibrillation is required.
It is well accepted that paravalvular leak and regurgitation following TAVI procedure have a negative prognostic effect.11 Recent advances in TAVI have been designed to reduce paravalvular leak as much as possible, with the majority of devices using some form of adaptive seal or skirt to accommodate irregularities in the annulus. There is, however, a trade-off between annular sealing and disruption of the conduction system among current-generation valves. Therefore, previously quoted PPI rates from an era where a degree of regurgitation was acceptable may not be applicable to modern practice. This is especially true in patients with a more oval-shaped annulus, where imposing a circular device will cause more distortion and require greater radial force. Further investigation into PPI rates and the degree of circularity is required.
These data were obtained using the original Lotus valve, since which time the design of the valve has been modified, as has the implantation technique, in an effort to reduce interaction with the membranous ventricular septum. There would be an expectation that current pacing rates would be lower than those seen in this study.
There are some data suggesting that PPI may be protective against sudden cardiac death in patients post TAVI who have a new onset left bundle-branch block.12 Although the exact cause of death (ventricular arrhythmia vs atrioventricular block) has not been established, necrosis of the His-Purkinje fibers and atrioventricular node in these patients at post mortem supports the atrioventricular block hypothesis. It is interesting to note that there were no differences in left bundle-branch block rates between the two groups in this study.
Study limitations. It is important to acknowledge some limitations of this study. First, we did not have ambulatory monitoring on the patients before TAVI and it may be that some had occult or transient conduction disease, which was uncovered by admission for the procedure. Choice of valve size and implantation depth are known to be important factors in determining the need for PPI and, as this is a single-center study, other operators may have different PPI rates. Pressure on bed occupancy encouraged a low threshold for pacing in our center.
The need for PPI is a common consequence of TAVI. In patients receiving a Lotus device, there is very low risk of PPI requirement if the PR interval on day 1 is <230 ms, regardless of QRS duration. Those >230 ms require monitoring until day 3, by which time those requiring PPI have “declared” themselves by rising PR interval or higher degrees of atrioventricular block. This may facilitate early discharge, which is of increasing significance as TAVI is extended into a lower-risk cohort.
From the Sussex Cardiac Centre, Brighton and Sussex University Hospitals, Brighton, United Kingdom.
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Cockburn reports grant support/proctor income from Boston Scientific. Dr Hildick-Smith reports grant support from Boston Scientific, Medtronic, and Edwards Lifesciences. The authors report no conflicts of interest regarding the content herein.
Manuscript submitted January 19, 2020, final version accepted January 29, 2020.
Address for correspondence: David Hildick-Smith, MD, Sussex Cardiac Centre, Brighton and Sussex University Hospitals, Eastern Road Brighton, BN2 5BE United Kingdom. Email: firstname.lastname@example.org
- Leon MB, Smith CR, Mack M, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010;363:1597-1607.
- Lev M. The normal anatomy of the conduction system in man and its pathology in atrioventricular block. Ann N Y Acad Sci. 1964;111:817-829.
- Erdogan HB, Kayalar N, Ardal H, et al. Risk factors for requirement of permanent pacemaker implantation after aortic valve replacement. J Card Surg. 2006;21:211-215.
- Harris A, Davies M, Redwood D, Leatham A, Siddons H. Aetiology of chronic heart block. A clinico-pathological correlation in 65 cases. Br Heart J. 1969;31:206-218.
- Keefe DL, Griffin JC, Harrison DC, Stinson EB. Atrioventricular conduction abnormalities in patients undergoing isolated aortic or mitral valve replacement. Pacing Clin Electrophysiol. 1985;8:393-398.
- Totaro P, Calamai G, Montesi G, Barzaghi C, Vaccari M. Continuous suture technique and impairment of the atrioventricular conduction after aortic valve replacement. J Card Surg. 2000;15:418-422.
- Barbanti M, Petronio AS, Ettori F, et al. 5-year outcomes after transcatheter aortic valve implantation with CoreValve prosthesis. JACC Cardiovasc Interv. 2015;8:1084-1091.
- Schwerg M, Fulde F, Dreger H, Poller WC, Stangl K, Laule M. Optimized implantation height of the Edwards SAPIEN 3 valve to minimize pacemaker implantation after TAVI. J Interv Cardiol. 2016;29:370-374. Epub 2016 May 31.
- De Torres-Alba F, Kaleschke G, Diller GP, et al. Changes in the pacemaker rate after transition from Edwards SAPIEN XT to SAPIEN 3 transcatheter aortic valve implantation: the critical role of valve implantation height. JACC Cardiovasc Interv. 2016;9:805-813.
- Meredith IT, Walters DL, Dumonteil N, et al. 1-year outcomes with the fully repositionable and retrievable Lotus transcatheter aortic replacement valve in 120 high-risk surgical patients with severe aortic stenosis: results of the REPRISE II study. JACC Cardiovasc Interv. 2016;9:376-384.
- Abdel-Wahab M, Zahn R, Horack M, et al. Aortic regurgitation after transcatheter aortic valve implantation: incidence and early outcome. Results from the German transcatheter aortic valve interventions registry. Heart. 2011;97:899-906.
- Urena M, Webb JG, Eltchaninoff H, et al. Late cardiac death in patients undergoing transcatheter aortic valve replacement: incidence and predictors of advanced heart failure and sudden cardiac death. J Am Coll Cardiol. 2015;65:437-448.