Transcatheter aortic valve implantation (TAVI) is now considered standard treatment for patients with severe symptomatic aortic stenosis (AS) in whom the risks of surgical aortic valve replacement (AVR) outweigh the benefits. Moreover, TAVI is now an acceptable alternative to surgery for high-risk patients.1,2 The use of this technology has been growing at a fast rate since the first implantation, which was performed by Cribier et al3 in 2002, using a balloon-expandable stent and equine valve. It is well documented that TAVI improves New York Heart Association functional class in patients with severe AS.4,5 The PARTNER trial has demonstrated enhanced survival of 20% in absolute terms at 1 year, in inoperable patients treated with TAVI compared with medical therapy including balloon valvuloplasty.1 For patients deemed high risk for surgery, there was no significant difference in survival between TAVI and AVR.2 However, TAVI is associated with a relatively high early and midterm mortality. One-year mortality in PARTNER trial cohorts A and B was 24.2% and 30.7%, respectively. Large registries have shown 30-day mortality of approximately 10% with varying rate of complications. Previous studies have shown 1-year survival rates ranging from 12% to 60%, allowing for learning curve experiences.1,2,6–9
The short- and long-term survival after conventional AVR are excellent even in high-risk populations. The logistic European System for Cardiac Operative Risk Evaluation (EuroSCORE) overestimates the risks associated with surgery and should be used in caution in deciding the best modality of treatment for patients with severe AS.10 The UK TAVI Registry reports that poor left ventricular (LV) function, presence of moderate/severe aortic regurgitation, and chronic obstructive pulmonary disease are independent predictors of mortality after TAVI.11 Transcatheter aortic valve implantation–induced left bundle branch block (LBBB) has been shown to be an independent predictor of mortality.12 One study has reported that cardiac comorbidities (low-gradient AS, tricuspid valve regurgitation) are associated with all-cause mortality at 6-month follow-up.13 Muñoz-García et al14 have demonstrated that early mortality beyond 30 days is predicted by preoperative comorbidity scores and the functional status of the patient. A pooled analysis of 12 studies and 1223 TAVI patients looking at the perioperative causes of mortality showed that these were widely variable and of both cardiac and noncardiac origin.15 Identifying the predictive factors and causes of mortality after TAVI is important to improve patient selection process and to reduce associated serious complications.
The objectives of this study were to determine the short-term and the midterm survival after TAVI, to determine the predictive factors of mortality, and to identify the actual causes of death during follow-up.
PATIENTS AND METHODS
Between December 2007 and May 2012, a total of 119 patients with symptomatic severe AS underwent 121 TAVI procedures with either the Medtronic CoreValve (Medtronic, Inc, Minneapolis, MN USA) or the Edwards SAPIEN (Edwards Lifesciences, Irvine, CA USA). The decision for a TAVI procedure was made in a multidisciplinary meeting that consisted of at least two cardiac surgeons, one interventional cardiologist, one noninterventional cardiologist, one cardiac anesthetist-intensivist, one neurologist, and one pulmonary physician. The patients with a EuroSCORE of 12 or greater, and/or in whom the operative risks were thought to be higher than 15%, were discussed at the multidisciplinary meeting for consideration of TAVI. The decision of the access route was determined by the multidisciplinary team, with the transfemoral approach as the first choice.
All patients underwent a systematic process of clinical evaluation and angiographic and echocardiographic assessments. The TAVI imaging consisted of peripheral angiography, detailed measurement of the aortic root using transthoracic and transesophageal echocardiography, aortography, and coronary angiography. Where indicated, computed tomographic and magnetic resonance image scanning and lung function tests were also performed. Hospital mortality, transient ischemic attack or stroke, peripheral vessel complications, wound infection, renal replacement therapy requiring hemofiltration, myocardial infarction, pacemaker implantation, blood transfusion, platelet transfusion, prolonged ventilation defined as greater than 24 hours, and intensive care and hospital stay were recorded.
The patients underwent transfemoral, transaortic, or subclavian CoreValve (CoreValve Revalving System; Medtronic, Inc, Minneapolis, MN USA) or transapical SAPIEN valve (Cribier-Edwards/Edwards SAPIEN THV; Edwards Lifesciences, Irvine, CA USA) insertion. Patients were selected for femoral or nonfemoral approach depending on the condition of the iliofemoral arteries. All procedures apart from one were performed under general anesthesia. The patients received aspirin and clopidogrel preoperatively. During the procedure, unfractioned heparin was administred to achieve an activated clotting time of 250 seconds.
Follow-up was completed using data from bereavement’s office, coroner’s office, outpatient clinics, referral hospitals, and general practitioners. Follow-up was completed in all patients. Death was defined as mortality from all causes after valve insertion.
Univariate analysis of dichotomous, categorical, and continuous preintervention and postintervention data were carried out to determine their influence or relationship with postintervention mortality.
Data were tabulated as 2 × 2, or contingency tables. Pearson χ2 and Fisher exact tests were used to compare groups as alive or dead. The distribution of continuous variables was assessed for normality with the Shapiro-Wilk test, and Mann-Whitney U tests were used to compare both groups.
Several models of multivariate regression analyses were made with all independent categorical and continuous variables that appeared significant (P < 0.20) for postintervention mortality, and using backward stepwise (likelihood ratio) Cox regression analysis, the most significant variables that predicted postintervention mortality over time were determined. Kaplan-Meier survival analysis was performed to calculate mean survival time, numbers at risk per year, and survival curve.
The mean ± SD age was 81 ± 9 years, and 59% were men. The mean ± SD EuroSCORE was 22 ± 15. A total of 119 patients with severe AS who underwent 121 TAVI procedures were included. The median follow-up time was 1.29 years (range, 0–4.5). Seventy-five patients (63%) were in New York Heart Association functional class III to IV. Significant concomitant coronary artery disease and prior cardiac surgery were present in 29 (24%) and 52 (43%) patients, respectively. The total number of deaths was 36 (30%), with 5 deaths (4.2%) that occurred within 30 days. Baseline characteristics and risk factors are shown in Table 1.
The implantation was successful in 92% of the cases, and procedural mortality was 0.8% (one patient). The transfemoral approach was used in 76% of the patients. One hundred ten patients (91%) had CoreValve prosthesis, and 11 (9%) had Edwards SAPIEN valves. One patient had cardiac arrest due to left coronary occlusion immediately after the implantation of the valve and was successfully converted to AVR. The patient made a very good recovery and was discharged at 10 days after surgery. Two patients required second TAVI at a later date after the original implantation to correct severe paravalvular aortic regurgitation. The incidence of vascular injury and stroke was 2.4% and 4.1%, respectively.
Intraprocedural death occurred in one patient. One patient developed pericardial tamponade that required surgical intervention.
After successful TAVI, 21 patients (17.6%) had new conduction abnormalities that required permanent pacemaker implantation before hospital discharge. Paravalvular aortic regurgitation was moderate in 10 patients (8%) and mild in 47 patients (39%). Sixty-four patients (53%) had none or trace aortic regurgitation after TAVI.
Two patients had myocardial infarction. In one patient, the valve was deployed and migrated into the LV cavity. The patient was transferred to the theater for emergency AVR. Perioperatively, he sustained a major stroke. Transient ischemic attacks or aortic dissection did not occur in any of the study patients. Acute kidney injury occurred in 12.3% of the patients, none of whom required dialysis during hospitalization. Periprocedural outcomes are shown in Table 2.
In our study, a total of 36 patients (30.2%) died during follow-up at a median of 1.29 years. The overall actuarial survival rates at 30 days, 12 months, and 24 months were 95.8%, 83.2%, and 76.5%, respectively. The Kaplan-Meier survival curve for the population in our study is shown in Figure 1. Our study identified several significant predictors of overall mortality in the whole population: history of atrial fibrillation (P = 0.03), history of heart block (P < 0.01), LV dysfunction (P = 0.04), and critical preoperative state (P < 0.01).
Causes of mortality after TAVI were widely variable and of both cardiac and noncardiac nature. When a death occurred within the first 30 days, it was mainly cardiac in nature (80%). One patient died because of cardiac tamponade after removal of the temporary pacing wires. Another patient died because of multiorgan failure after early removal of the pacing electrode, with occurrence of conduction disturbances and cardiac arrest. In 12 (33%) of the 36 patients who died, mortality was related to cardiac causes. Twelve (33%) of the 36 deaths were due to bronchopneumonia. Twelve patients (34%) died because of a variety of other reasons such as pulmonary embolism, stroke, cancer, renal failure, and sepsis. Causes of death are listed in Table 3.
The principal findings of the present study are encouraging results of short-term to midterm survival after TAVI. However, LV dysfunction, prior atrial fibrillation, and prior heart block seem to be independent predictive factors of all-cause mortality.
Several national registries have established an acceptable early mortality after TAVI ranging from 5% to 10.4%.1,2,6–9 In our study, the 30-day mortality was 4.2%. We believe that the keys for this finding were careful patient selection and optimum perioperative management. It is worth noting that four deaths (80%) within the first 30 days were cardiac related. Two patients died after complications related to manipulation of pacing wires. This necessitated the need for change in protocol for placement of temporary pacing wires and their removal after the procedure. In our TAVI cohort, we report good survival rates at 1 and 2 years of 83.2% and 76.5%, respectively. These data compare favorably with results after AVR in octogenarians and high-risk subgroups.10 Doss and colleagues16 reported an overall mortality of 13% after a mean ± SD follow-up of 3.8 ± 2 years after TAVI, with no valve degeneration.
Consistent with previous studies, we found impaired LV function to be independently associated with worse outcomes after TAVI. This finding provides some support that LV should be given close attention and optimization where possible in preoperative assessment of TAVI patients. Two other independent predictors of mortality were prior atrial fibrillation (P = 0.03) and prior heart block (P < 0.01). In our study, 19 patients (15.9%) had a history of atrial fibrillation before the procedure, and 9 (47.3%) of them died during follow-up. The possible impact of atrial fibrillation on mortality after TAVI may have been underestimated. It seems to play an important role in the thromboembolic events after TAVI. The optimum antithrombotic therapy in TAVI patients is still unclear. Prior atrioventricular block (LBBB, right bundle branch block, and first-degree heart block) was present in eight patients (6.7%), with five deaths (62.5%) in this group. The incidence of LBBB after TAVI ranges between 7% and 83% depending on the type of prosthesis used.12 Piazza and colleagues17 have reported that preexisting right bundle branch block is an independent risk factor of the development of complete heart block after TAVI. Furthermore, Houthuizen and associates12 have demonstrated that LBBB after TAVI is associated with significantly higher all-cause mortality.
Cardiac arrest, heart failure, and pneumonia were the causes of deaths within the first 30 days. Most of the late deaths were caused by noncardiac reasons, with bronchopneumonia being reported as the most common cause of late mortality. Applying all means necessary to prevent hospital- or community-acquired pneumonia in this high-risk population is important.
Manipulation of the temporary pacing wires in two patients led to serious complications and subsequent deaths. This finding highlights the need for better strategies in pacemaker management after TAVI. It is worth noting that four patients (11%) died of cancer during follow-up. This fact supports the need for comprehensive preoperative assessment of the elderly patients referred for TAVI.
Careful patient selection process, a multidisciplinary team approach, and good perioperative and postoperative management are essential for good outcomes. Greater involvement of physicians for the care of the elderly in the multidisciplinary team, comprehensive cancer screening of patients, and specific algorithms to estimate life expectancy can improve identification of patients with reasonable life expectancy who will benefit from TAVI.
This study reflects a single-center experience, and the sample size is small. Most of the patients received CoreValve prosthesis, and the results may be affected by this uneven distribution.
Short-term to midterm survival after TAVI was promising. Prior atrial fibrillation, heart block, and LV impairment were the most independent predictive factors of mortality. Thirty-day mortality was mainly due to cardiac reasons. The most common cause of late mortality was bronchopneumonia. Elderly patients referred for TAVI should be targeted with comprehensive preoperative assessment to identify those with life expectancy of greater than 1 year.
1. Leon MB, Smith CR, Mack M, et al. for the PARTNER Trial Investigators. Transcatheter aortic-valve implantation for aortic stenosis
in patients who cannot undergo surgery. N Engl J Med
. 2010; 363: 1597–1607.
2. Smith CR, Leon MB, Mack MJ, et al. for the PARTNER Trial Investigators. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med
. 2011; 364: 2187–2198.
3. Cribier A, Eltchaninoff H, Bash A, et al. Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis
: first human case description. Circulation
. 2002; 106: 3006–3008.
4. Webb JG, Pasupati S, Humphries K, et al. Percutaneous transarterial aortic valve replacement
in selected high-risk patients with aortic stenosis
. 2007; 116: 755–763.
5. Grube E, Schuler G, Buellesfeld L, et al. Percutaneous aortic valve replacement
for severe aortic stenosis
in high-risk patients using the second- and current third-generation self-expanding CoreValve prosthesis: device success and 30-day clinical outcome. J Am Coll Cardiol
. 2007; 50: 69–76.
6. Eltchaninoff H, Prat A, Gilard M, et al., for the FRANCE Registry Investigators. Transcatheter aortic valve implantation
: early results of the FRANCE (FRench Aortic National CoreValve and Edwards) registry. Eur Heart J
. 2011; 32: 191–197.
7. Rodés-Cabau J, Dumont E, De LaRochellière R, et al. Feasibility and initial results of percutaneous aortic valve implantation including selection of the transfemoral or transapical approach in patients with severe aortic stenosis
. Am J Cardiol
. 2008; 102: 1240–1246.
8. Tamburino C, Capodanno D, Ramondo A, et al. Incidence and predictors of early and late mortality after transcatheter aortic valve implantation
in 663 patients with severe aortic stenosis
. 2011; 123: 299–308.
9. Zahn R, Gerckens U, Grube E, et al., for the German Transcatheter Aortic Valve Interventions-Registry Investigators. Transcatheter aortic valve implantation
: first results from a multi-centre real-world registry. Eur Heart J
. 2011; 32: 198–204.
10. Mølstad P, Veel T, Rynning S. Long-term survival after aortic valve replacement
in octogenarians and high-risk subgroups. Eur J Cardiothorac Surg
. 2012; 42: 934–940.
11. Moat NE, Ludman P, de Belder MA, et al. Long-term outcomes after transcatheter aortic valve implantation
in high-risk patients with severe aortic stenosis
: the U.K. TAVI (United Kingdom Transcatheter Aortic Valve Implantation
) Registry. J Am Coll Cardiol
. 2011; 58: 2130–2138.
12. Houthuizen P, Van Garsse LA, Poels TT, et al. Left bundle-branch block induced by transcatheter aortic valve implantation
increases risk of death. Circulation
. 2012; 126: 720–728.
13. Gotzmann M, Pljakic A, Bojara W, et al. Transcatheter aortic valve implantation
in patients with severe symptomatic aortic valve stenosis—predictors of mortality and poor treatment response. Am Heart J
. 2011; 162: 238–245.
14. Muñoz-García AJ, Hernández-García JM, Jiménez-Navarro MF, et al. Survival and predictive factors of mortality after 30 days in patients treated with percutaneous implantation of the CoreValve aortic prosthesis. Am Heart J
. 2012; 163: 288–294.
15. Moreno R, Calvo L, Salinas P, et al. Causes of peri-operative mortality after transcatheter aortic valve implantation
: a pooled analysis of 12 studies and 1223 patients. J Invasive Cardiol
. 2011; 23: 180–184.
16. Doss M, Buhr EB, Martens S, Moritz A, Zierer A. Transcatheter-based aortic valve implantations at midterm: what happened to our initial patients? Ann Thorac Surg
. 2012; 94: 1400–1406.
17. Piazza N, Onuma Y, Jesserun E, et al. Early and persistent intraventricular conduction abnormalities and requirements for pacemaking after percutaneous replacement of the aortic valve. J Am Coll Cardiol
. 2008; 1: 310–316.
This report from Dr Alassar and the group at St. George’s Hospital in London examined predictive factors and causes of mortality after transcatheter aortic valve replacement (TAVR) at early and midterm follow-up. The transfemoral approach was used in three quarters of their patients. One hundred ten patients had the Medtronic CoreValve prosthesis, and 11 had an Edwards SAPIEN valve. Follow-up was completed for all patients at a median of 1.3 years. The 30-day mortality was 4.2%. Actuarial survival at 1, 2, and 3 years was 83%, 77%, and 68%, respectively. Survival was adversely affected by preoperative left ventricular dysfunction, atrial fibrillation, a history of prior heart block, and critical preoperative state. One-third of the deaths were due to pneumonia and one-third were due to cardiac causes. Although early mortality was mainly cardiac in origin, most of the late deaths were caused by noncardiac reasons.
This report of TAVR outcomes from an experienced center is informative. Both 30-day mortality and late survival were encouraging. The authors identified a number of predictive factors for mortality that may be helpful in the preoperative assessment of patients for this technology. The main limitation of the study is that it represents a single-center experience with a relatively small sample size and is not powered to adequately define all predictors of mortality.