Author + information
- Received August 10, 2007
- Revision received December 10, 2007
- Accepted December 20, 2007
- Published online June 1, 2008.
- Ajay Sinhal, MD,
- Lukas Altwegg, MD,
- Sanjeevan Pasupati, MBChB,
- Karin H. Humphries, DSc,
- Michael Allard, MD,
- Paul Martin, MBChBAO, PhD,
- Anson Cheung, MD,
- Jian Ye, MD,
- Charles Kerr, MD,
- Sam V. Lichtenstein, MD, PhD and
- John G. Webb, MD, FACC⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. John G. Webb, McLeod Professor of Heart Valve Intervention, St. Paul's Hospital, Room 476A, 1081 Burrard Street, Vancouver, British Columbia V6Z 1Y6, Canada.
Objectives Transcatheter aortic valve replacement (AVR) is a promising approach to aortic valve disease. The implications of this new therapy are not entirely known. We describe the potential for the development of new atrioventricular (AV) block.
Background Atrioventricular block is a known complication of conventional surgical AVR. Block is presumed to occur as a consequence of surgical trauma to the cardiac conduction tissue during excision of the diseased aortic valve and débridement of the calcified annulus. Whether AV block might occur as a consequence of nonsurgical implantation of an aortic stent valve is unknown.
Methods We reviewed our experience with patients undergoing transcatheter AVR using both the percutaneous transarterial and the open-chest direct left ventricular apical ventriculotomy approaches. Patients were considered at high risk for conventional surgery because of comorbidities. Continuous arrhythmia monitoring was performed for at least 48 h after the valve implantation procedure. Patients who developed apparently new, clinically significant AV block were identified.
Results Transcatheter AVR was successfully performed in 123 patients. Seventeen of these patients (13.8%) had pre-existing permanent pacemakers. Two patients (1.6%) required pacemaker implantation because of pre-existing intermittent bradycardia. Seven patients (5.7%) developed new and sustained complete AV block requiring pacemaker implantation. An additional 4 patients (3.3%) developed new and sustained left bundle branch block but did not require pacemaker implantation.
Conclusions As with conventional AVR surgery, transcatheter AVR may result in impaired atrioventricular conduction. Physicians and patients should be aware of the potential for AV block and pacemaker dependence.
Atrioventricular (AV) block is a well-described complication of surgical aortic valve replacement (AVR). New AV block reportedly requires pacemaker implantation in up to 6% of surgical patients (1–3). The mechanism is presumably injury to the cardiac conduction system during surgical excision of the adjacent diseased valve and annular tissue.
Transcatheter AVR is a relatively new alternative to conventional surgical valve replacement (Fig. 1) (4,5). In contrast to surgery, transcatheter AVR does not involve excision of the diseased native valve or annular tissue. Whether AV block can occur with transcatheter AVR because of annular dilatation and stent implantation in the absence of surgical excision of valve or annulus tissue is not known.
The procedure was approved for clinical use by the Department of Health and Welfare, Ottawa, Canada, in patients with severe, symptomatic aortic stenosis. Acceptance for the procedure required a consensus agreement among a group of senior surgeons and cardiologists that the risk of mortality or morbidity with conventional surgery was excessive caused by comorbidities. Patient or physician preference alone was not considered adequate (6). Written informed consent was obtained. All patients undergoing transcatheter AVR at our institution were prospectively followed up.
Transcatheter AVR was performed using either percutaneous femoral arterial access (7,8) or open-chest left ventricular puncture (5,9). Procedures were performed with fluoroscopic imaging and, for the most part, under a general anesthetic with transesophageal echocardiographic imaging. The diameter of the aortic annulus at the site of leaflet insertion was estimated in a long-axis view of the left ventricular outflow tract using transesophageal echocardiography.
The aortic valve was initially dilated using a standard valvuloplasty balloon with a nominal diameter approximately the same as the aortic annulus diameter as measured by echocardiography. A balloon-expandable valve (Cribier-Edwards or Edwards-SAPIEN, Edwards Lifesciences, Irvine, California) was used. This consists of a balloon-expandable stainless steel stent with an attached pericardial valve and fabric sealing cuff. Two sizes of prosthetic valve were used; a smaller prosthesis intended to be expanded with a 22-mm-diameter balloon to achieve a 23-mm external diameter and a larger prosthesis intended to be expanded with a 25-mm balloon to achieve a 26-mm external diameter. The 23-mm external diameter valve was considered suitable for an annulus diameter of 18 to 22 mm and a 26-mm valve for an annulus diameter of to 21 to 25 mm. Our approach was to routinely select a prosthesis that exceeded the measured annulus diameter by 10% to 20%. Routine oversizing was intended to securely fix the prosthesis within the native valve and annulus and minimize the potential for paravalvular leaks between the prosthetic valve and native annulus (4).
Burst rapid pacing at 150 to 220 beats/min was used to reduce cardiac motion and transvalvular flow during balloon dilation and prosthetic valve deployment (10). A temporary transvenous right ventricular lead was used for the percutaneous transarterial approach, and a left ventricular epicardial lead for the transapical approach. The temporary pacemaker leads were removed after valve implantation, unless required because of pacemaker dependence. Cardiopulmonary bypass was not used.
Patients were assessed with a history, electrocardiogram, and routine blood tests during screening, immediately before the procedure, daily for 3 days post-procedure, and at 30 days after the procedure. A transthoracic echocardiogram was obtained before and after valve implantation and at 30 days after the procedure. Electrocardiographic monitoring was performed during the procedure and continued for at least 48 h. In addition to interrogation of an ongoing prospective database, patient records were reviewed for prior evidence of atrioventricular block.
Continuous variables are presented as means or medians, as appropriate, or proportions for categorical variables. Given the small sample size, the Fisher exact test was used to compare proportions and the Mann-Whitney U test was used to compare continuous variables. Univariate logistic regression was used to examine potential factors associated with pacemaker requirement.
Transcatheter aortic valve implantation was successfully performed in 123 patients. Patients were generally elderly with multiple comorbidities (Table 1). The logistic EuroSCORE operative mortality estimate was 30.1% (range 19.5% to 42.8%) (11). Of the cohort of 123 patients, 17 (13.8%) were pacemaker dependent before aortic valve implantation. Characteristics of 106 patients without a pre-existing pacemaker are shown in Table 1. Two of 106 patients (1.9%) underwent post-procedural pacemaker implantation because of pre-existing episodic bradycardia. It seemed that episodic bradycardia was not new, but rather was only fully appreciated by in-hospital monitoring. In 7 of 106 patients (6.6%), new and complete AV block was evident immediately after transcatheter aortic valve implantation. Patients with heart block were monitored in the coronary care unit, and medications known to impair AV conduction were discontinued. Permanent pacemakers were implanted after 45 ± 23 h of continued dependency on temporary pacing.
New left bundle branch block was documented at the time of the post-procedural electrocardiogram in 7 patients (5.7%). Isolated left bundle branch block was transient in 3 (2.4%) but persisted until hospital discharge in 4 (3.2%). At 6 months follow-up, 6 of the 7 patients with transient or sustained isolated left bundle branch block remain well. One patient with transient left bundle branch block died of progressive congestive heart failure 4 months after transapical AVR. None have required pacemaker implantation.
At 6 months, 6 of 7 patients with new complete heart block remain well, but pacer dependent. In the remaining patient, persistent complete AV block immediately following valve implantation had developed and the patient underwent permanent pacemaker implantation on day 6. An acutely ischemic bowel resulted in death on day 10. At post-mortem, localized macroscopic myocardial necrosis was observed in the leftward basal interventricular septum (Fig. 1A). Necrosis was located slightly apical to the implant rather than directly adjacent to it. Microscopically, the necrosis showed a geographic pattern characteristic of ischemic injury with sparing of subendocardial myocytes. In addition, myocytes were replaced at the periphery of the injured area by a cellular infiltrate comprised of macrophages and mesenchymal cells, indicative of a reparative response and consistent with an onset of injury approximately 10 days previously. Myocardial injury was also seen microscopically in the uppermost portion of the leftward interventricular septum (Figs. 1B and 1C) in the immediate vicinity of the conducting tissues. The atrioventricular node itself did not show convincing evidence of myocyte injury. In the most superior aspects of the septum, myocytes in the subendocardial region were injured, whereas more inferiorly, but still deep to the stented prosthesis, the subendocardial myocytes were spared. No thrombi or emboli were noted in any of the intramyocardial arteries.
In the cohort of 123 patients, 17 had pre-existing permanent pacemakers and 7 required pacemakers after aortic valve implantation. Among those 99 patients who were not paced before or after valve implantation, pre-existing conduction abnormalities were common, being identified in 37 patients (37.3%) as shown in Table 1. These included first-degree block in 15 patients (15.2%), right bundle branch block in 9 (9.1%), left bundle branch block in 12 (12.1%), and second-degree heart block in 1 (1.0%). One of 7 patients developing new and sustained complete AV block had pre-existing right bundle branch block. No pre-existing conduction abnormalities were apparent in the remaining 6 individuals who developed new complete AV block. Univariate logistic regression analysis did not identify any risk factors associated with pacemaker requirement (Table 1).
New AV block is a known complication of surgical AVR, requiring pacemaker implantation in up to 6% of surgical patients (1–3). In our transcatheter AVR experience in high-risk patients with multiple comorbidies, a comparable number, 5.7% of all patients, required a permanent pacemaker because of apparently new AV block. To put this in context, AV block has also been described as a consequence of aortic balloon valvuloplasty (12), aortic root abscess (13), and percutaneous device closure of membranous ventricular septal defect (14). That AV block might occur because of trauma to ventricular conduction tissues adjacent to the aortic valve is perhaps not surprising given the close proximity of these 2 structures.
Reported risk factors for complete AV block after surgery include previous aortic regurgitation, myocardial infarction, pulmonary hypertension, and postoperative electrolyte imbalance (15). Koplan et al (15) reported that right bundle branch block was the strongest predictor of pacemaker dependency after surgical aortic valve replacement. Similarly Ben Ameur et al (16) noted that bifasicular block was associated with the need for post-operative pacing. Univariate analysis of our experience did not suggest risk factors for heart block, although patient and event numbers may have been insufficient.
In our series, 1 of 10 patients with pre-existing right bundle branch block required pacemaker implantation. An additional 4 patients with new sustained left bundle branch block in the absence of pre-existing right bundle branch block remain well. It seems reasonable to hypothesize that balloon and stent trauma in the region of the aortic annulus and left ventricular outflow tract affecting the adjacent left bundle branch would be more problematic in the presence of pre-existing compromise of the right bundle. A 10% incidence of complete block in the setting of pre-existing right bundle branch block would be consistent with this.
In our patients, the prosthetic valve was routinely dilated to a diameter slightly larger than the echocardiographically estimated annulus diameter. The intention was to securely fix the prosthesis within the annulus and minimize paravalvular regurgitation. Although not confirmed by univariate analysis, it seems reasonable to speculate that the use of relatively larger valve sizes and greater degrees of prosthesis/annulus mismatch might result in greater compression of adjacent structures and more likely result in impaired AV conduction.
We attempted to distinguish between new bradycardia as a consequence of AVR and pre-existing bradycardia unrelated to AVR. However, pre-existing episodic bradycardia may have been undetected in some patients, leading to overestimation of the risk attributable to the procedure, particularly given the frequency of comorbidities and the lack of a comparator group. Larger numbers of patients would be required to adequately assess risk factors for heart block. There are considerable differences between various types of implantable valves, and it is possible that the likelihood of AV block may vary.
As with conventional surgery, new or worsened AV block may occur as a consequence of transcatheter aortic valve implantation. Physicians and patients should be aware of the potential for AV block and pacemaker dependence.
Dr. John Webb is a consultant to Edwards Lifesciences, Irvine, California.
- Abbreviations and Acronyms
- aortic valve replacement
- left ventricular ejection fraction
- New York Heart Association
- Received August 10, 2007.
- Revision received December 10, 2007.
- Accepted December 20, 2007.
- American College of Cardiology Foundation
- Limongelli G.,
- Ducceschi V.,
- D'Andrea A.,
- et al.
- Kolh P.,
- Lahaye L.,
- Gerard P.,
- Limet R.
- Webb J.G.,
- Chandavimol M.,
- Thompson C.R.,
- et al.
- Vassiliades T.A. Jr..,
- Block P.C.,
- Cohn L.H.,
- et al.
- Webb J.G.,
- Chandavimol M.,
- Thompson C.R.,
- et al.
- Webb J.G.,
- Pasupati S.J.,
- Humphries K.,
- et al.
- Lichtenstein S.V.,
- Cheung A.,
- Ye J.,
- et al.
- Nashef S.A.,
- Roques F.,
- Hammill B.G.,
- et al.
- Gougeon F.
- Koplan B.A.,
- Stevenson W.G.,
- Epstein L.M.,
- Aranki S.F.,
- Maisel W.H.