**Transcatheter Aortic Valve Implantation: State of the Art**

Alice Le Huu, Rony Atoui and Dominique Shum-Tim *Divisions of Cardiac Surgery and Surgical Research, Department of Surgery, McGill University Health Center, Montreal, Canada* 

## **1. Introduction**

210 Aortic Valve

Varadarajan P., Kapoor N., Bansal R.C., *et al.* (2006). Clinical profile and natural history of

Walther T., Simon P., Dewey T., *et al.* (2007). Transapical minimally invasive aortic valve implantation: multicenter experience, *Circulation* 116(Suppl 11): I240-I245. Walther T., Dewey T., Borger M.A., *et al.* (2009). Transapical aortic valve implantation: step

Webb J.G., Pasupati S., Humphries K., *et al.* (2007). Percutaneous transarterial aortic valve

Wenaweser P, Buellesfeld L, Gerckens U, Grube E. (2007). Percutaneous aortic valve

Zegdi R., Ciobotaru V., Noghin M., *et al.* (2008). Is it reasonable to treat all calcified stenotic

82(6): 2111-2115.

*Interv* 70(5) :760-764

*Am Coll Cardiol* 51(5): 579-584.

755-763.

by step, *Ann Thorac Surg* 87(1): 276-283.

453 nonsurgical managed patients with severe aortic stenosis, *Ann Thorac Surg* 

replacement in selected high-risk patients with aortic stenosis, *Circulation* 116(7):

replacement for sévère aortic régurgitation in degenerated bioprosthesis : the first valve in valve procédure using Corevlave Revalving system, *Catheter Cardiovasc* 

aortic valves with a valved stent? Results from a human anatomic study in adults, *J* 

In the last century, medical innovation has revolutionized human lives and the management of medical diseases. Conditions which were once considered untreatable are now managed and even cured. Consequently, life expectancy has dramatically increased. The aging population brings about new challenges and pathologies that must be addressed with different approaches. As the body ages, so does the heart, bringing aortic stenosis to the forefront of valvular heart disease. It is estimated that 4.6% of patients over the age of 75 years old suffer from aortic stenosis (Nkomo et al., 2006). Traditionally, there were three modalities of treatment: conservative medical management, balloon valvuloplasty, and surgical aortic valve replacement (AVR). Long considered to be the gold-standard for aortic valve stenosis, surgical intervention provided a functional valve with acceptable mortality rates. However, the risks of surgical intervention increase dramatically depending on a patient's comorbidities. Accordingly, high-risk patients were often relegated to medical management or balloon valvuloplasty. Conservative management has yielded extremely disappointing results. Patients who underwent balloon valvuloplasty in conjunction with medical treatment had a 44-37.2% mortality rate within a year. In addition, conservative management is associated with a high rate of restenosis. Medical management alone resulted in an unacceptable 25% mortality in one year (Nkomo et al., 2006; Ben-Dor et al., 2010). Despite these staggering statistics, the Euro Heart Survey suggested that approximately 30% of patients suffering from severe aortic stenosis were not treated with surgical intervention (Lung et al., 2003). A significant portion of these patients are refused surgery because they are deemed to have elevated surgical risks. Considering the natural history of the pathology, patients who were refused for surgery suffer considerable morbidity and mortality. Even with maximum medical therapy, the future for these patients appeared bleak.

Fortunately, the advent of transcatheter aortic valve implantation (TAVI), has led to new options for non-surgical candidates. As early as 1965, an article describing a cathetermounted valve replacement for temporary relief of aortic insufficiency in an animal (Davies H., 1965). After more than three decades of development, Cribier et al. (2002) successfully implanted a percutaneous prosthetic heart valve in a 57-year-old man. The patient, who was moribund due to numerous medical conditions, showed significant clinical and echocardiographic improvement after the valve had been implanted. Although he eventually

process.

active endocarditis.

for implantation.

**3. Design of the device** 

**4. Methods of approach** 

26-mm valve, respectively (Lange et al., 2007).

Transcatheter Aortic Valve Implantation: State of the Art 213

pelvis are done to delineate the anatomy of the patient. For technical considerations, the presence of torturous, severely diseased femoral or subclavian arteries may not be amenable to percutaneous TAVI. When available, some centers use three-dimensional reconstruction of the aortic root to detect calcifications, femoral stenosis, kinking, and aortic dissection (Geisbusch et al., 2010). While helpful, reconstructions are not essential to the screening

Absolute contraindications to TAVI are few, but include intolerance of anticoagulation and bleeding diathesis. There is a lack of consensus on the ideal strategy, but many centers use dual antiplatelet therapy: a loading dose of aspirin (325 mg) and clopidogrel (300-600 mg), followed by clopidogrel (75 mg) for 6 months and aspirin (81 mg) indefinitely (Gurvitch et al., 2010). Other exclusion criteria include the presence of a ventricular or atrial clot and

Currently, two types of percutaneous valves dominate the market: the SAPIEN Valve manufactured by Edwards Life Sciences (Irvine, CA, USA) and the CoreValve produced by Medtronic, Inc. (Minneapolis, MN, USA). Each valve has unique properties and indications

The SAPIEN model is a trileaflet bovine valve mounted on a cobalt-chromium stent; it is a balloon-expandable prosthesis. Designed for antegrade, retrograde, or transapical delivery, the valve is deployed by balloon expansion into a subcoronary position. The prosthetic valve uses the native calcified valve to anchor itself (Eltchaninoff et al., 2008). There are three sizes available: 23 mm, 26 mm, and 29 mm, with a 20-mm size in development (Bande et al., 2010). Presently, the SAPIEN valve employs the Novoflex delivery system, which allows the valve to be delivered via an 18F and 19F introducer sheath for a 23-mm and

The Medtronic CoreValve, is used worldwide but has yet to be approved in North America and Japan. The CoreValve is a trileaflet bioprosthetic porcine valve which is mounted onto a self-expanding nitinol frame. The nitinol stent has 3 portions: an upper portion which serves to anchor the device to the ascending aorta, a central component of the stent engineered to avoid obstructing the coronary ostium, and a lower portion of the stent which rests on the annulus. The CoreValve system is believed to adapt to non-circular anatomies. Two sizes are available: 22 mm and 29 mm; the valve requires an 18F introducer sheath (Bande et al., 2010). Designed for retrograde delivery, it is implanted using a transfemoral or subclavian

Presently, three approaches for TAVI are used: transfemoral, transapical, and subclavian. Selecting the proper approach involves integrating patient characteristics with technical considerations. The preferred access remains the transfemoral approach. Using this access, the Novaflex delivery system requires arteries to be 6.0 to 6.5 mm in diameter. Similarly, the CoreValve requires 6.0 to 7.0 mm arteries in order to accommodate the catheter-mounted aortic valve. The size of the introducer sheath used is dictated by patient anatomy, primarily the size of the femoral artery for access and the diameter of the valve required to fit the annulus. Depending on the sheath used, surgical exposure and closure of the femoral artery

approach; however, successful transapical implantation has been reported.

succumbed to his multiple comorbidities, his last transesophageal echocardiography demonstrated a functional aortic valve (Webb et al., 2009). A landmark success, this intervention was followed by a flurry of developments which resulted in considerable refinement of the procedure. This would culminate in increased safety, efficiency, and physician familiarity with the intervention. It is now estimated that over 15,000 patients worldwide have undergone a TAVI procedure (Geisbusch et al., 2010).

## **2. Patient selection**

With any medical procedure, whether it is aortic valve replacement or medical treatment, the key to a favorable outcome is appropriate patient selection. By virtue of its recent development, access to TAVI remains restricted. Presently, TAVI is offered only to patients with symptomatic, critical aortic stenosis who have been deemed unsuitable for AVR. Evaluating a patient as inoperable depends on many factors, which include patient comorbidities, the surgeon's experience, and the institution in which the surgeon practices. Subtle details may influence a physician's judgment, which renders it difficult to provide a standard definition of a non-surgical candidate. With this limitation in mind, many institutions qualify patients with a logistic EuroScore calculated ≥ 20% or an STS (Society of Thoracic Surgery) predicted mortality risk score ≥ 10% as high-risk (Bande et al., 2010). Although not included in these two scoring systems, other criteria often cited when deeming a patient high-risk include: calcified porcelain aorta, chest wall deformities, cancer, cirrhosis with portal hypertension, neurological dysfunction, perceived frailty, severe chronic obstructive disease, previous cardiac surgery, severe cerebrovascular disease, low ejection fraction, and untreatable coronary artery disease (Saia et al., 2010). Traditionally, a bicuspid valve is considered a contraindication to TAVI. Due to its elliptical rather than circular shape, TAVI may result in a morphologically distorted valve and increased incidence of perivalvular leaks. Consequently, valve durability may be compromised. Recently published experience with 11 bicuspid TAVI demonstrated decreased gradients across the valve, and increased valve area. Two patients had moderate leaks. Although interesting, these results did not address the fundamental concern of TAVI in bicuspid valves related to their durability (Wijesinghe et al., 2010).

Once patients are deemed inoperable, they must be carefully screened before proceeding to TAVI. In many centers, candidates are evaluated by a multidisciplinary committee consisting of cardiologists, cardiac surgeons, and cardiac anesthesiologists. Therefore, patients are evaluated on a case-by-case basis. Broad exclusion criteria may include dementia, life expectancy of less than 1 year, severely incapacitating neurological dysfunction, thoracic aneurysm, and a low likelihood of benefitting from the procedure (Shareghi et al., 2007). However, these selection parameters are subject to the discretion of each individual committee, and remain obscure. As TAVI becomes more prevalent, it will most certainly entail the development of specific guidelines to help orient physicians.

A patient's anatomy is crucial in the selection process. The preoperative screening tests are similar in many centers. Transthoracic echocardiography and a coronary angiogram are standard. From a purely logistic point of view, a patient with and aortic annulus diameter < 18 mm or > 26 mm may be excluded due to limited availability of valve sizes (Webb et al., 2009). However, this exclusion is likely to be attenuated in the coming years as a wider range of valve sizes are developed. Computer tomography of the thorax, abdomen, and pelvis are done to delineate the anatomy of the patient. For technical considerations, the presence of torturous, severely diseased femoral or subclavian arteries may not be amenable to percutaneous TAVI. When available, some centers use three-dimensional reconstruction of the aortic root to detect calcifications, femoral stenosis, kinking, and aortic dissection (Geisbusch et al., 2010). While helpful, reconstructions are not essential to the screening process.

Absolute contraindications to TAVI are few, but include intolerance of anticoagulation and bleeding diathesis. There is a lack of consensus on the ideal strategy, but many centers use dual antiplatelet therapy: a loading dose of aspirin (325 mg) and clopidogrel (300-600 mg), followed by clopidogrel (75 mg) for 6 months and aspirin (81 mg) indefinitely (Gurvitch et al., 2010). Other exclusion criteria include the presence of a ventricular or atrial clot and active endocarditis.

## **3. Design of the device**

212 Aortic Valve

succumbed to his multiple comorbidities, his last transesophageal echocardiography demonstrated a functional aortic valve (Webb et al., 2009). A landmark success, this intervention was followed by a flurry of developments which resulted in considerable refinement of the procedure. This would culminate in increased safety, efficiency, and physician familiarity with the intervention. It is now estimated that over 15,000 patients

With any medical procedure, whether it is aortic valve replacement or medical treatment, the key to a favorable outcome is appropriate patient selection. By virtue of its recent development, access to TAVI remains restricted. Presently, TAVI is offered only to patients with symptomatic, critical aortic stenosis who have been deemed unsuitable for AVR. Evaluating a patient as inoperable depends on many factors, which include patient comorbidities, the surgeon's experience, and the institution in which the surgeon practices. Subtle details may influence a physician's judgment, which renders it difficult to provide a standard definition of a non-surgical candidate. With this limitation in mind, many institutions qualify patients with a logistic EuroScore calculated ≥ 20% or an STS (Society of Thoracic Surgery) predicted mortality risk score ≥ 10% as high-risk (Bande et al., 2010). Although not included in these two scoring systems, other criteria often cited when deeming a patient high-risk include: calcified porcelain aorta, chest wall deformities, cancer, cirrhosis with portal hypertension, neurological dysfunction, perceived frailty, severe chronic obstructive disease, previous cardiac surgery, severe cerebrovascular disease, low ejection fraction, and untreatable coronary artery disease (Saia et al., 2010). Traditionally, a bicuspid valve is considered a contraindication to TAVI. Due to its elliptical rather than circular shape, TAVI may result in a morphologically distorted valve and increased incidence of perivalvular leaks. Consequently, valve durability may be compromised. Recently published experience with 11 bicuspid TAVI demonstrated decreased gradients across the valve, and increased valve area. Two patients had moderate leaks. Although interesting, these results did not address the fundamental concern of TAVI in bicuspid

Once patients are deemed inoperable, they must be carefully screened before proceeding to TAVI. In many centers, candidates are evaluated by a multidisciplinary committee consisting of cardiologists, cardiac surgeons, and cardiac anesthesiologists. Therefore, patients are evaluated on a case-by-case basis. Broad exclusion criteria may include dementia, life expectancy of less than 1 year, severely incapacitating neurological dysfunction, thoracic aneurysm, and a low likelihood of benefitting from the procedure (Shareghi et al., 2007). However, these selection parameters are subject to the discretion of each individual committee, and remain obscure. As TAVI becomes more prevalent, it will most certainly

A patient's anatomy is crucial in the selection process. The preoperative screening tests are similar in many centers. Transthoracic echocardiography and a coronary angiogram are standard. From a purely logistic point of view, a patient with and aortic annulus diameter < 18 mm or > 26 mm may be excluded due to limited availability of valve sizes (Webb et al., 2009). However, this exclusion is likely to be attenuated in the coming years as a wider range of valve sizes are developed. Computer tomography of the thorax, abdomen, and

worldwide have undergone a TAVI procedure (Geisbusch et al., 2010).

valves related to their durability (Wijesinghe et al., 2010).

entail the development of specific guidelines to help orient physicians.

**2. Patient selection** 

Currently, two types of percutaneous valves dominate the market: the SAPIEN Valve manufactured by Edwards Life Sciences (Irvine, CA, USA) and the CoreValve produced by Medtronic, Inc. (Minneapolis, MN, USA). Each valve has unique properties and indications for implantation.

The SAPIEN model is a trileaflet bovine valve mounted on a cobalt-chromium stent; it is a balloon-expandable prosthesis. Designed for antegrade, retrograde, or transapical delivery, the valve is deployed by balloon expansion into a subcoronary position. The prosthetic valve uses the native calcified valve to anchor itself (Eltchaninoff et al., 2008). There are three sizes available: 23 mm, 26 mm, and 29 mm, with a 20-mm size in development (Bande et al., 2010). Presently, the SAPIEN valve employs the Novoflex delivery system, which allows the valve to be delivered via an 18F and 19F introducer sheath for a 23-mm and 26-mm valve, respectively (Lange et al., 2007).

The Medtronic CoreValve, is used worldwide but has yet to be approved in North America and Japan. The CoreValve is a trileaflet bioprosthetic porcine valve which is mounted onto a self-expanding nitinol frame. The nitinol stent has 3 portions: an upper portion which serves to anchor the device to the ascending aorta, a central component of the stent engineered to avoid obstructing the coronary ostium, and a lower portion of the stent which rests on the annulus. The CoreValve system is believed to adapt to non-circular anatomies. Two sizes are available: 22 mm and 29 mm; the valve requires an 18F introducer sheath (Bande et al., 2010). Designed for retrograde delivery, it is implanted using a transfemoral or subclavian approach; however, successful transapical implantation has been reported.

## **4. Methods of approach**

Presently, three approaches for TAVI are used: transfemoral, transapical, and subclavian. Selecting the proper approach involves integrating patient characteristics with technical considerations. The preferred access remains the transfemoral approach. Using this access, the Novaflex delivery system requires arteries to be 6.0 to 6.5 mm in diameter. Similarly, the CoreValve requires 6.0 to 7.0 mm arteries in order to accommodate the catheter-mounted aortic valve. The size of the introducer sheath used is dictated by patient anatomy, primarily the size of the femoral artery for access and the diameter of the valve required to fit the annulus. Depending on the sheath used, surgical exposure and closure of the femoral artery

Transcatheter Aortic Valve Implantation: State of the Art 215

Comparisons between the three methods of delivery remain in their infancy. The studies have used small sample sizes, and do not offer direct comparisons between the methods of delivery, nor were the patients randomized between the methods of approach. The trend is to favor a transfemoral approach, and, when this is not possible, to opt for a transapical approach. In theory, a transapical approach requires mechanical ventilation and general

Initial results from TAVI are very encouraging, albeit of a small scale. However, in recent years, large-scale studies support the assertion that TAVI is a viable treatment option for aortic stenosis. The most important study to date is the PARTNER trial, a North American multicenter trial with 2 main cohorts, run by Leon et al. (2010). Surgical candidates were randomized to TAVI or conventional aortic valve replacement (Cohort A). Patients who were deemed non-operable (358 patients) were randomized to TAVI and medical therapy (including balloon valvuloplasty: cohort B). The patients all underwent TAVI through a transfemoral approach, and were followed for up to one year post-procedure. The results of cohort B were recently published. TAVI was shown to be superior for all of the hard endpoints, including death at one year: 30.7% for TAVI compared with 71.6% for standard therapy. Patients who underwent the procedure demonstrated significantly increased freedom from symptoms associated with their valvular disease: 74.8% compared with 42.0% (Leon et al., 2010). This study is unequivocal proof that TAVI is superior to conventional medical management. The results from this randomized trial are in accordance with other studies. Gurvitch et al. (2010), sought to elucidate the long-term outcome of TAVI patients by following a cohort of 70 individuals over a period of 3 years. Survival after 1 year (excluding 11 patients that died within 30 days of the procedure and 8 patients with failed valve implantation) was 81%. Studies with mixed delivery approaches (transfemoral, transapical, and subclavian), yielded mortality rates between 5.4%-11.3% after 30 days (Gurvitch et al., 2010; Petronio et al., 2010). However, with cohort studies, it is difficult to compare the mortality rate with conventional medical treatment. Indirect comparisons can be made with published estimates of mortality, which ranges from 51%-62% in symptomatic patients after 1 year (Bach et al., 2007). The totality of the data suggests that in patients who

Aside from mortality and freedom from symptoms, another important determinant of outcome is the durability of the transcatheter valves. Studies indicate that the TAVI device is quite durable. Gurvitch et al. (2010) suggests that the valve appears to be structurally normal, with very little incidence of degeneration at 3 years. Valve function appears to decrease over the years, quantified by a slight decrease in calculated valve area and an increase in valve gradient, estimated by the authors at 0.06% cm2 per year. It is unclear whether these changes will be clinically significant. Due to its relatively recent development, the durability of the valve will be known only in the near future. It is important to note that surgically replaced aortic valves also show signs of degeneration, tearing, and calcification. The durability of surgically replaced valves compared to TAVI will only be known in years

The presence of aortic insufficiency after TAVI is another important criterion of success. Unfortunately, this endpoint is exceedingly difficult to qualify because of the subjective

anesthesia, which renders it less desirable.

are refused surgery, TAVI is superior to medical management.

**5. Outcome** 

to come.

may be needed. In other cases, a smaller catheter may require only a vascular closure device, and thus negate the need of a surgical team (Eltchaninoff et al., 2008). A heavily calcified arterial system with numerous tortuosities may not be amenable to access. In addition, patients who have had extensive vascular surgery are usually not suited to this approach. The transfemoral is considered to be a relatively safe method of access. However, reported complications include formation of retroperitoneal hematoma, iliac or femoral artery pseudoaneurysms, and stenosis or occlusion induced by a vascular closure device. According to Philipp et al. (2009), the majority of these complications are managed conservatively, or with endovascular intervention.

Typically, the transapical approach will be used when the iliofemoral approach is not feasible. For this approach, the SAPIEN valve is used because it is approved for antegrade delivery. The procedure is done in a hybrid suite with the guidance of transesophageal echocardiography and fluoroscopy. Under, under general anesthesia, a 6 to 8 cm anterolateral thoracotomy incision is made to expose the pericardium. Once open, temporary pacing wires are attached and a suitable puncture site in the left ventricular apex is identified. Two orthogonal U-shaped pledgeted sutures are placed in the myocardium before a needle puncture is done, followed by the insertion of a guide wire to the ascending aorta. The catheter is then increased in size to allow for balloon valvuloplasty. Once complete, the sheath size is increased to accommodate the delivery system, and the transcatheter valve is deployed. At the end of the procedure, at the discretion of the surgeon, a drain may be left in place. This method of delivery is much more invasive than the iliofemoral approach and is linked to significant complications. This may include pericarditis, tamponade, pneumothorax, and the formation of a false aneurysm at the apex of the heart (Al-Attar et al., 2009). Injury to the mitral valve using this approach is also a particular concern due to its anatomic proximity. In addition, recovery time is more extensive. Pain at the thoracotomy site is not negligible and can result in respiratory debilitation in frail and elderly patients, culminating in prolonged hospitalization.

Recently, the trans-subclavian approach has been developed as an alternative to the transapical approach. A technique described by Petronio et al. (2010) consists of surgical exposure of the subclavian artery, followed by arterial puncture between purse-string sutures. A 6F catheter is inserted, and the 18F catheter commonly used for the transfemoral approach is placed over a guide wire. It is then threaded through the subclavian artery and into the ascending aorta. Once in place, the catheter mounted valve is implanted using the same method as for the transfemoral approach. Concerns exist over the security of the transsubclavian method. Of particular interest are patients who have had their internal mammary used in a coronary artery bypass graft. It is unclear whether this approach can safely preserve flow through the graft during implantation of the valve. By rapidly withdrawing the 18F catheter, the Petronio et al. (2010) team was able to avoid all signs of myocardial ischemia. Caution must be exercised in the interpretation of this data, as the sample size (8 patients) was quite small. Other considerations concerning the transsubclavian approach include direct injury to the brachial plexus and damage secondary to hematoma formation. Perhaps because of its recent development, this method appears to be used less frequently than the transfemoral approach. It is usually considered a second option when the transfemoral anatomy is unsuitable for access. As the number of TAVI procedures increase, familiarity with the trans-subclavian approach will rise, and it will become more frequently employed.

Comparisons between the three methods of delivery remain in their infancy. The studies have used small sample sizes, and do not offer direct comparisons between the methods of delivery, nor were the patients randomized between the methods of approach. The trend is to favor a transfemoral approach, and, when this is not possible, to opt for a transapical approach. In theory, a transapical approach requires mechanical ventilation and general anesthesia, which renders it less desirable.

## **5. Outcome**

214 Aortic Valve

may be needed. In other cases, a smaller catheter may require only a vascular closure device, and thus negate the need of a surgical team (Eltchaninoff et al., 2008). A heavily calcified arterial system with numerous tortuosities may not be amenable to access. In addition, patients who have had extensive vascular surgery are usually not suited to this approach. The transfemoral is considered to be a relatively safe method of access. However, reported complications include formation of retroperitoneal hematoma, iliac or femoral artery pseudoaneurysms, and stenosis or occlusion induced by a vascular closure device. According to Philipp et al. (2009), the majority of these complications are managed conservatively, or

Typically, the transapical approach will be used when the iliofemoral approach is not feasible. For this approach, the SAPIEN valve is used because it is approved for antegrade delivery. The procedure is done in a hybrid suite with the guidance of transesophageal echocardiography and fluoroscopy. Under, under general anesthesia, a 6 to 8 cm anterolateral thoracotomy incision is made to expose the pericardium. Once open, temporary pacing wires are attached and a suitable puncture site in the left ventricular apex is identified. Two orthogonal U-shaped pledgeted sutures are placed in the myocardium before a needle puncture is done, followed by the insertion of a guide wire to the ascending aorta. The catheter is then increased in size to allow for balloon valvuloplasty. Once complete, the sheath size is increased to accommodate the delivery system, and the transcatheter valve is deployed. At the end of the procedure, at the discretion of the surgeon, a drain may be left in place. This method of delivery is much more invasive than the iliofemoral approach and is linked to significant complications. This may include pericarditis, tamponade, pneumothorax, and the formation of a false aneurysm at the apex of the heart (Al-Attar et al., 2009). Injury to the mitral valve using this approach is also a particular concern due to its anatomic proximity. In addition, recovery time is more extensive. Pain at the thoracotomy site is not negligible and can result in respiratory debilitation in frail and elderly patients,

Recently, the trans-subclavian approach has been developed as an alternative to the transapical approach. A technique described by Petronio et al. (2010) consists of surgical exposure of the subclavian artery, followed by arterial puncture between purse-string sutures. A 6F catheter is inserted, and the 18F catheter commonly used for the transfemoral approach is placed over a guide wire. It is then threaded through the subclavian artery and into the ascending aorta. Once in place, the catheter mounted valve is implanted using the same method as for the transfemoral approach. Concerns exist over the security of the transsubclavian method. Of particular interest are patients who have had their internal mammary used in a coronary artery bypass graft. It is unclear whether this approach can safely preserve flow through the graft during implantation of the valve. By rapidly withdrawing the 18F catheter, the Petronio et al. (2010) team was able to avoid all signs of myocardial ischemia. Caution must be exercised in the interpretation of this data, as the sample size (8 patients) was quite small. Other considerations concerning the transsubclavian approach include direct injury to the brachial plexus and damage secondary to hematoma formation. Perhaps because of its recent development, this method appears to be used less frequently than the transfemoral approach. It is usually considered a second option when the transfemoral anatomy is unsuitable for access. As the number of TAVI procedures increase, familiarity with the trans-subclavian approach will rise, and it will

with endovascular intervention.

culminating in prolonged hospitalization.

become more frequently employed.

Initial results from TAVI are very encouraging, albeit of a small scale. However, in recent years, large-scale studies support the assertion that TAVI is a viable treatment option for aortic stenosis. The most important study to date is the PARTNER trial, a North American multicenter trial with 2 main cohorts, run by Leon et al. (2010). Surgical candidates were randomized to TAVI or conventional aortic valve replacement (Cohort A). Patients who were deemed non-operable (358 patients) were randomized to TAVI and medical therapy (including balloon valvuloplasty: cohort B). The patients all underwent TAVI through a transfemoral approach, and were followed for up to one year post-procedure. The results of cohort B were recently published. TAVI was shown to be superior for all of the hard endpoints, including death at one year: 30.7% for TAVI compared with 71.6% for standard therapy. Patients who underwent the procedure demonstrated significantly increased freedom from symptoms associated with their valvular disease: 74.8% compared with 42.0% (Leon et al., 2010). This study is unequivocal proof that TAVI is superior to conventional medical management. The results from this randomized trial are in accordance with other studies. Gurvitch et al. (2010), sought to elucidate the long-term outcome of TAVI patients by following a cohort of 70 individuals over a period of 3 years. Survival after 1 year (excluding 11 patients that died within 30 days of the procedure and 8 patients with failed valve implantation) was 81%. Studies with mixed delivery approaches (transfemoral, transapical, and subclavian), yielded mortality rates between 5.4%-11.3% after 30 days (Gurvitch et al., 2010; Petronio et al., 2010). However, with cohort studies, it is difficult to compare the mortality rate with conventional medical treatment. Indirect comparisons can be made with published estimates of mortality, which ranges from 51%-62% in symptomatic patients after 1 year (Bach et al., 2007). The totality of the data suggests that in patients who are refused surgery, TAVI is superior to medical management.

Aside from mortality and freedom from symptoms, another important determinant of outcome is the durability of the transcatheter valves. Studies indicate that the TAVI device is quite durable. Gurvitch et al. (2010) suggests that the valve appears to be structurally normal, with very little incidence of degeneration at 3 years. Valve function appears to decrease over the years, quantified by a slight decrease in calculated valve area and an increase in valve gradient, estimated by the authors at 0.06% cm2 per year. It is unclear whether these changes will be clinically significant. Due to its relatively recent development, the durability of the valve will be known only in the near future. It is important to note that surgically replaced aortic valves also show signs of degeneration, tearing, and calcification. The durability of surgically replaced valves compared to TAVI will only be known in years to come.

The presence of aortic insufficiency after TAVI is another important criterion of success. Unfortunately, this endpoint is exceedingly difficult to qualify because of the subjective

Transcatheter Aortic Valve Implantation: State of the Art 217

dislocation during implantation, while Rhodes-Cabu et al. (2010) had a rate of 2%. Management of this complication primarily consists of retrieving the valve in the ascending aorta and reimplanting it. However, patients with this complication had significantly higher rates of stroke, ischemia, or renal failure than did patients whose first implantation was successful (Geisbusch et al., 2010). Although this complication is documented in other papers, this study on the CoreValve reveals it to be quite frequent. It is probable that valve dislocation is a common occurrence; however, it is not well documented in large trials as a potential source of morbidity. In addition, in cohort studies which involve databases, this

Other complications include vascular-related injuries as such as dissection, perforations, ruptures, and hematomas, as well as cardiac-related issues such as aortic root rupture,

Preliminary data on transcatheter aortic valve implantation is very promising. It also highlights the importance of a dedicated multidisciplinary team consisting of a cardiac surgeon, interventional cardiologist, echocardiographer, cardiac anesthesiologist, perfusionist, and scrub team. As with any new technology, more research needs to be completed before its role in the management of aortic stenosis can be clearly delineated. At present, it is clearly superior to medical management, and offers non-surgical patients a viable option. However, its use in patients who are surgical candidates is likely to be tempered by several limiting factors. Of primary importance is the increased risk of cerebral ischemia during this procedure, that seems to persist overtime even after the placement of TAVI. By virtue of its delivery system and implantation mechanism, it will be very difficult to minimize thromboembolic events. In older, frail patients with many co-morbidities, TAVI may be an acceptable alternative. New generation of sutureless surgically implantable prosthetic valves are another surgical options in development (Perseval S valve by Sorin). These valves are designed to be implanted after the calcified aortic valve is surgically removed. This will theoretically diminish the potential of ongoing embolic risks that were observed in current TAVI reported by the PARTNER trial (Miller, 2011). Without the need to suture the prosthetic valve onto the aortic annulus, the valve would be quicker to implant and result in decreased cardio-pulmonary bypass time (O'Brien et al., 1998). Consequently, these sutureless valves may be more suitable for patients with intermediate risk profile, in whom a shorter cardio-pulmonary bypass time may be beneficial. Ongoing accumulation of experience and longer term follow-up are underway to assess the clinical application of this new generation of surgically implantable sutureless valve in many cardiac centers. In a younger patient, however, where mortality associated with surgical aortic valve replacement is very low,

Another area of interest is the durability of TAVI. The valve's ability to expand to fit the aortic root may be detrimental to its long-term use. It remains questionable whether this type of valve is suitable for implantation when the patient's life expectancy is beyond ten to

Currently, many questions remain unanswered. The TAVI technique is still in its infancy compared with traditional surgical and medical management of aortic stenosis, which has undergone rigorous investigation. For surgical valve replacement, there are clear indications

fifteen years. Unfortunately, this data will be available only several years from now.

information is more difficult to obtain, and is thus excluded from analysis.

mitral valve injury, tamponade, and injury to the left ventricle apex.

**7. Future prospects** 

traditional surgical management is indicated.

nature of echographic findings. Notoriously operator-dependent, this issue is magnified when exams are completed by different physicians and technicians in a range of institutions. Nevertheless, the PARTNER trial found moderate or severe paravalvular aortic insufficiency in 11.8% of patients after 30 days, and 10.5% at 1 year (Leon et al., 2010). The amount of aortic regurgitation remained stable in patients after one year, did not seem to progress, and was generally well tolerated by the patients. Similarly, Gurvitch et al. (2010) determined that 6% of their patients had moderate aortic regurgitation, with no incidences of severe regurgitation. The clinical significance of these echographic findings is yet to be determined. The stability of the aortic insufficiency over time will be the key determinant of the relevance of this finding.

## **6. Complications**

As with any other medical procedure, there are risks associated with TAVI. The most concerning complication is the risk of stroke. The culprits behind the neurological event are multifold. Atheroma in the aorta, dislodged during TAVI, may play a significant role. In addition, during TAVI many patients undergo balloon valvuloplasty, which is likely to dislodge microemboli of calcium to the cerebrovascular system. The TAVI procedure, with the self-expanding valve or manual positioning of the valve, is also traumatic to the original valve. This provides another possible source of emboli. Despite encouraging results in terms of crude outcomes such as death and freedom from repeat hospitalization, the optimism for this procedure is tempered by the risk of cerebral ischemia. The PARTNER trial found the incidence of stroke to be 5% compared with 1.1% in the medically treated group (Leon et al., 2010). Similarly, Gurvitch et al. (2010) documented a risk of 8.6% after 3 years. Even more alarming, Kahlert et al. (2010) followed 32 patients who underwent TAVI (either the SAPIEN or CoreValve), to discover the effect on their neurological status. Neurological assessment and cognitive function were assessed through the National Institute Health Stroke Scale (NIHSS) along with a Mini Mental State Examination (MMSE). In conjunction with these clinical exams, patients received preoperative and postoperative MRIs. In 84% of the patients, significant multiple and diffuse emboli were discovered on the MRIs. The investigators were unable to find a clinical correlation between the radiologic images and patients' symptoms. However, the NIHSS and the MMSE are crude methods of measuring cognitive function. Subtle changes in memory and cognitive function may be much more evident in a younger, healthier individual with fewer comorbidities. Further research must be completed on the neurological impact of TAVI before it can be offered to patients who are surgical candidates.

The need for pacemaker insertion following TAVI is also well documented. A study by Khawaja et al. (2011) quoted a rate of 33.3% requiring pacemaker insertion in the 30 days following the procedure. The study found that patients were most often susceptible to developing left bundle branch block. After studying the CoreValve, the authors reasoned that the skirt of the valve (depending on the position after the deployment) will lie in close proximity to the left bundle branch, which entails conduction abnormalities. The old valve is not excised and the placement of the CoreValve on top of the remnants may lead to compression of conduction structures.

Dislocation or migration of the valve during implantation is mentioned in several case reports. In a patient population of 212, Geisbusch et al. (2010) reported an incidence of 10% dislocation during implantation, while Rhodes-Cabu et al. (2010) had a rate of 2%. Management of this complication primarily consists of retrieving the valve in the ascending aorta and reimplanting it. However, patients with this complication had significantly higher rates of stroke, ischemia, or renal failure than did patients whose first implantation was successful (Geisbusch et al., 2010). Although this complication is documented in other papers, this study on the CoreValve reveals it to be quite frequent. It is probable that valve dislocation is a common occurrence; however, it is not well documented in large trials as a potential source of morbidity. In addition, in cohort studies which involve databases, this information is more difficult to obtain, and is thus excluded from analysis.

Other complications include vascular-related injuries as such as dissection, perforations, ruptures, and hematomas, as well as cardiac-related issues such as aortic root rupture, mitral valve injury, tamponade, and injury to the left ventricle apex.

## **7. Future prospects**

216 Aortic Valve

nature of echographic findings. Notoriously operator-dependent, this issue is magnified when exams are completed by different physicians and technicians in a range of institutions. Nevertheless, the PARTNER trial found moderate or severe paravalvular aortic insufficiency in 11.8% of patients after 30 days, and 10.5% at 1 year (Leon et al., 2010). The amount of aortic regurgitation remained stable in patients after one year, did not seem to progress, and was generally well tolerated by the patients. Similarly, Gurvitch et al. (2010) determined that 6% of their patients had moderate aortic regurgitation, with no incidences of severe regurgitation. The clinical significance of these echographic findings is yet to be determined. The stability of the aortic insufficiency over time will be the key determinant of

As with any other medical procedure, there are risks associated with TAVI. The most concerning complication is the risk of stroke. The culprits behind the neurological event are multifold. Atheroma in the aorta, dislodged during TAVI, may play a significant role. In addition, during TAVI many patients undergo balloon valvuloplasty, which is likely to dislodge microemboli of calcium to the cerebrovascular system. The TAVI procedure, with the self-expanding valve or manual positioning of the valve, is also traumatic to the original valve. This provides another possible source of emboli. Despite encouraging results in terms of crude outcomes such as death and freedom from repeat hospitalization, the optimism for this procedure is tempered by the risk of cerebral ischemia. The PARTNER trial found the incidence of stroke to be 5% compared with 1.1% in the medically treated group (Leon et al., 2010). Similarly, Gurvitch et al. (2010) documented a risk of 8.6% after 3 years. Even more alarming, Kahlert et al. (2010) followed 32 patients who underwent TAVI (either the SAPIEN or CoreValve), to discover the effect on their neurological status. Neurological assessment and cognitive function were assessed through the National Institute Health Stroke Scale (NIHSS) along with a Mini Mental State Examination (MMSE). In conjunction with these clinical exams, patients received preoperative and postoperative MRIs. In 84% of the patients, significant multiple and diffuse emboli were discovered on the MRIs. The investigators were unable to find a clinical correlation between the radiologic images and patients' symptoms. However, the NIHSS and the MMSE are crude methods of measuring cognitive function. Subtle changes in memory and cognitive function may be much more evident in a younger, healthier individual with fewer comorbidities. Further research must be completed on the neurological impact of TAVI before it can be offered to patients who

The need for pacemaker insertion following TAVI is also well documented. A study by Khawaja et al. (2011) quoted a rate of 33.3% requiring pacemaker insertion in the 30 days following the procedure. The study found that patients were most often susceptible to developing left bundle branch block. After studying the CoreValve, the authors reasoned that the skirt of the valve (depending on the position after the deployment) will lie in close proximity to the left bundle branch, which entails conduction abnormalities. The old valve is not excised and the placement of the CoreValve on top of the remnants may lead to

Dislocation or migration of the valve during implantation is mentioned in several case reports. In a patient population of 212, Geisbusch et al. (2010) reported an incidence of 10%

the relevance of this finding.

**6. Complications** 

are surgical candidates.

compression of conduction structures.

Preliminary data on transcatheter aortic valve implantation is very promising. It also highlights the importance of a dedicated multidisciplinary team consisting of a cardiac surgeon, interventional cardiologist, echocardiographer, cardiac anesthesiologist, perfusionist, and scrub team. As with any new technology, more research needs to be completed before its role in the management of aortic stenosis can be clearly delineated. At present, it is clearly superior to medical management, and offers non-surgical patients a viable option. However, its use in patients who are surgical candidates is likely to be tempered by several limiting factors. Of primary importance is the increased risk of cerebral ischemia during this procedure, that seems to persist overtime even after the placement of TAVI. By virtue of its delivery system and implantation mechanism, it will be very difficult to minimize thromboembolic events. In older, frail patients with many co-morbidities, TAVI may be an acceptable alternative. New generation of sutureless surgically implantable prosthetic valves are another surgical options in development (Perseval S valve by Sorin). These valves are designed to be implanted after the calcified aortic valve is surgically removed. This will theoretically diminish the potential of ongoing embolic risks that were observed in current TAVI reported by the PARTNER trial (Miller, 2011). Without the need to suture the prosthetic valve onto the aortic annulus, the valve would be quicker to implant and result in decreased cardio-pulmonary bypass time (O'Brien et al., 1998). Consequently, these sutureless valves may be more suitable for patients with intermediate risk profile, in whom a shorter cardio-pulmonary bypass time may be beneficial. Ongoing accumulation of experience and longer term follow-up are underway to assess the clinical application of this new generation of surgically implantable sutureless valve in many cardiac centers. In a younger patient, however, where mortality associated with surgical aortic valve replacement is very low, traditional surgical management is indicated.

Another area of interest is the durability of TAVI. The valve's ability to expand to fit the aortic root may be detrimental to its long-term use. It remains questionable whether this type of valve is suitable for implantation when the patient's life expectancy is beyond ten to fifteen years. Unfortunately, this data will be available only several years from now.

Currently, many questions remain unanswered. The TAVI technique is still in its infancy compared with traditional surgical and medical management of aortic stenosis, which has undergone rigorous investigation. For surgical valve replacement, there are clear indications

Transcatheter Aortic Valve Implantation: State of the Art 219

Khawaja MZ, Rajani R, Cook A, Khavandi A, Moynagh A, Chowdhary S, Spence SM, Brown

Lange R, Schreiber C, Gotz W, Hettich I, Will A, Libera P, Laborde JC & Beurnschmitt R.

Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, Tuzcu EM, Webb JG,

Masson JB, Lee M, Boone RH, Al Ali A, Bugami SA, Hamburger J, Mancini J, Ye J, Cheung

O'Brien MF, Gardner MA, Garlick RB, Davison MB, Thomson HL, Burstow DJ. (1998). The

Petronio AS, De Carlo M, Bedogoni F, Marzocchi A, Klugmann S, Maisano F, Ramondo A,

Philipp K, Al-Rashid F, Weber M, Wendt D, Heine T, Kottenberg E, Thielmann M, Kühl H,

*Cardiovascular Medicine (Hagerstown)*. Vol. 10, 727-732 ISSN 1558-2027 Shareghi S, Rasouli L, Shavelle DM, Burstein S & Matthews RV. (2007). Current Results Of

*of Medicine*. Vol. 17, pp. 1597-607 ISSN 0028-4793

951-960 ISSN 0009-7322

1522-6662

Pennsylvania

Vol. 13, pp. 376–385 ISSN 0886-0440

Vol. 19, pp. S37-S42 ISSN 1042-3931

*Intervention*. Vol. 3, pp. 359-366 ISSN 1941-7640

S, Khan SQ, Walker N, Trivedi U, Hutchinson N, De Belder AJ, Moat N, Blackman DJ, Levy RD, Manoharan G, Roberts D, Khogali SS, Crean P, Brecker SJ, Baumbach A, Mullen M, Laborde JC & Hildick-Smith D. (2011). Permanent Pacemaker Insertion After CoreValve Transcatheter Aortic Valve Implantation: Incidence and Contributing Factors (the UK CoreValve Collaborative). *Circulation.* Vol. 123, pp.

(2007). First Successful Transapical Aortic Valve Implantation With the CoreValve Revalving System: A Case Report. *Heart Surgery Forum*. Vol. 10, E478-E479 ISSN

Fontana GP, Makkar RR, Brown DL, Block PC, Guyton RA, Pichard AD, Bavaria JE, Herrmann HC, Douglas PS, Petersen JL, Akin JJ, Anderson WN, Wang D & Pocock S; PARTNER Trial Investigators. (2010). Transcatheter Aortic-Valve Implantation For Aortic Stenosis In Patients Who Cannot Undergo Surgery. *New England Journal* 

A, Humphries KH, Wood J, Nietlispach F & Webb JG. (2010). Impact Of Coronary Artery Disease On Outcomes After Transcatheter Aortic Valve Implantation. *Catheterization and Cardiovascular Intervention.* Vol. 76, pp. 165-173 ISSN 1522-1946 Miller, C. (2011). "Late-breaking Clinical Trial: Frequency and Severity of Neurologic Events

after Aortic Valve Replacement or Transcatheter Aortic Valve Implantation in High Surgical Risk Patients with Aortic Stenosis: The PARTNER Trial Stroke Substudy," American Association for Thoracic Surgery 91st Annual Meeting, Philadelphia,

Cryolife-O'Brien stentless aortic porcine xenograft valve. *Journal of Cardiac Surgery*

Ussia GP, Ettori F, Poli A, Brambilla N, Saia F, De Marco F & Colombo A. (2010). Safety and Efficacy of the Subclavian Approach For the Transcatheter Aortic Valve Implantation with the CoreValve Revalving System. *Circulation: Cardiovascular* 

Peters J & Heinz JG. (2009). Vascular Access Site Complications After Percutaneous Transfemoral Aortic Valve Implantation. *Herz*. Vol. 5, pp. 398-408, ISSN 0340-9937 Saia F, Marrozzini C, Dall'Ara G, Russo V, Martin-Suarez S, Ortolani P, Palmerini T, Taglieri

N, Bordoni B, Pilato E, Di Bartolomeo R, Branzi A & Marzocchi A. (2010). How Many Patients With Severe Symptomatic Aortic Stenosis Excluded For Cardiac Surgery Are Eligible For Transcatheter Heart Valve Implantation? *Journal of* 

Balloon Aortic Valvuloplasty In High Risk Patients. *Journal of Invasive Cardiology*.

and guidelines on the timing and management of replacement. Data on proper management of TAVI patients are lacking, including basic premises such as anticoagulation and management in conjunction with other heart pathologies. Aortic stenosis rarely exists as a single entity, and ischemic heart disease often coexists with this pathology. Initial data on the role of concomitant coronary artery disease and TAVI is conflicting. Dewey et al. (2010) reported increased mortality in patients with coronary artery disease undergoing TAVI, while Masson et al. (2010) found that this factor did not affect mortality. These are conflicting reports and illustrate that the application of TAVI in complex clinical situations remains obscure. More research must be completed to properly define the role of TAVI in the treatment of aortic stenosis.

It is undeniable that the advent of TAVI has provided physicians with an important tool for treating aortic valve disease. In non-surgical, high-risk patients, the technology may mean freedom from considerable mortality and morbidity. The therapeutic potential of this technology is astonishing; however, expectations must be tempered by caution.

#### **8. References**


and guidelines on the timing and management of replacement. Data on proper management of TAVI patients are lacking, including basic premises such as anticoagulation and management in conjunction with other heart pathologies. Aortic stenosis rarely exists as a single entity, and ischemic heart disease often coexists with this pathology. Initial data on the role of concomitant coronary artery disease and TAVI is conflicting. Dewey et al. (2010) reported increased mortality in patients with coronary artery disease undergoing TAVI, while Masson et al. (2010) found that this factor did not affect mortality. These are conflicting reports and illustrate that the application of TAVI in complex clinical situations remains obscure. More research must be completed to properly define the role of TAVI in

It is undeniable that the advent of TAVI has provided physicians with an important tool for treating aortic valve disease. In non-surgical, high-risk patients, the technology may mean freedom from considerable mortality and morbidity. The therapeutic potential of this

Al-Attar N, Ghodbane W, Himbert D, Rau C, Raffoul R, Messika-Zeitoun D, Brochet E,

Dewey TM, Brown DL, Herbert MA, Culica D, Smith CR, Leon MB, Svensson LG Tuzcu M,

Geisbusch S, Bleiziffer S, Mazzitelli D, Ruge H, Bauernschmitt R &Lange R. (2010). Incidence

Gurvitch R, Wood DA, Tay EL, Leipsic J, Ye J, Lichtenstein SV, Thompson CR, Carere RG,

Kahlert P, Knipp SC, Schlamann M, Thielmann M, Al-Rashid F, Weber M, Johansson U,

Vahanian A & Nataf P. (2009) Unexpected Complications Of Transapical Aortic Valve Implantation. *Annals of Thoracic Surgery* Vol. 88, pp. 90-94, ISSN 0003-4975 Bach DS, Cimino N & Deeb MG. (2007) Unoperated Patients with Severe Aortic Stenosis*. Journal of the American College of Cardiology*. Vol. 50, pp. 2018-2019 ISSN 0735-1097 Bande M, Michev I, Sharp A, Chieffo A & Colombo A. (2010). Percutaneous Transcatheter

Aortic Valve Implantation: Past Accomplishments, Present Achievements and Applications, Future Perspectives. *Cardiology in Review*. Vol. 18, pp. 111-124, ISSN

Webb JG, Cribier A & Mack MJ. (2010). Effect Of Concomitant Coronary Artery Disease On Procedural and Late Outcomes Of Transcatheter Aortic Valve Implantation. *The Annals of Thoracic Surgery.* Vol. 89, pp. 758-767 ISSN 0003-4975 Eltchaninoff H, Zajarias A & Tron C. (2008). Transcatheter Aortic Valve Implantation:

Technical Aspects, Results and Indications. *Archives Cardiovascular Disease*. Vol. 101,

and Management of CoreValve Dislocation During Transcatheter Aortic Valve Implantation. *Circulation: Cardiovascular Intervention*. Vol. 3, pp. 531-536, ISSN 1941-

Wijesinghe N, Nietlispach F, Boone RH, Lauck S, Cheung A & Webb JG. (2010). Transcatheter Aortic Valve Implantation: Durability of Clinical and Hemodynamic Outcomes Beyond 3 Years in a Large Patient Cohort. *Circulation*. Vol. 122, pp. 1319-

Wendt D, Jakob HG, Forsting M, Sack S, Erbel R & Eggebrecht H. (2010). Silent and Apparent Cerebral Ischemia After Percutaneous Transfemoral Aortic Valve Implantation: A Diffuse-Weighted Magnetic Resonance Imaging Study. *Circulation*.

technology is astonishing; however, expectations must be tempered by caution.

the treatment of aortic stenosis.

**8. References** 

1061-5377

7640

pp. 126-132, ISSN 1875-2136

1327, ISSN 0009-7322

Vol. 121, pp. 870-878 ISSN 0009-7322


**1. Introduction** 

public health problem in the ensuing decades.

comorbidities and may facilitate faster recovery.

**12** 

*UK* 

*Wessex Cardiac Centre,* 

*Southampton University Hospitals Trust* 

**Transcatheter Aortic Valve Implantation** 

Calcific aortic stenosis (AS) is the most common form of degenerative heart valve disease in developed countries. AS predominantly affects the elderly with a prevalence of 4% in people over the age of 80 years. With increasing longevity and an ageing population the absolute number of people afflicted with AS is set to rise. The clinical course of the disease is insidious at first (Ross & Braunwald, 1968), but is followed by rapid progression once symptoms of congestive cardiac failure, angina and syncope develop (Cheitlin *et al.*, 1979; Otto *et al.*, 1989; Davies *et al.*, 1991; Peter *et al.*, 1993). If left untreated mortality exceeds 50% at two years (Kelly *et al.*, 1988; Turina *et al.*, 1987) and AS is, therefore, set to become a major

Aortic valve replacement (AVR) has been the gold standard intervention for AS for more than 40 years (Charlson *et al.*, 2006) and over 60,000 procedures are performed annually in the European Union. However, one third of patients are denied access to surgery, often due to their advanced age (Iung *et al.*, 2003, 2005). Percutaneous balloon valvuloplasty was heralded initially as a promising breakthrough for high-risk symptomatic patients with AS (Cribier *et al.*, 1986). However, long-term follow-up has yielded unacceptable rates of restenosis and poor event-free survival (Otto *et al.*, 1994; Lieberman *et al.*, 1995). Balloon valvuloplasty is now recommended only as a bridge to emergency AVR or in the palliation

Transcatheter aortic valve implantation (TAVI) has recently emerged as an effective therapeutic alternative to conventional AVR for high-risk AS patients. TAVI was developed initially in porcine models (Anderson *et al.*, 1992), but it took a decade for this technology to be translated to humans (Cribier *et al.*, 2002). Initially, an antegrade transseptal approach was used, but this has now been superseded by transapical (Ye *et al.*, 2006; Webb *et al.*, 2007) and retrograde percutaneous techniques (transarterial or transaxillary) (Webb *et al.*, 2006). The range of different approaches has increased the feasibility of TAVI in patients with relative contraindications, such as extensive peripheral vascular disease, porcelain aorta and thoracic radiotherapy. TAVI is also less invasive than open AVR and permits replacement of the native diseased valve in the beating heart without the need for sternotomy and cardiopulmonary bypass. Consequently, TAVI may be less influenced by a patient's

A multidisciplinary team consisting of interventional cardiologists, cardiothoracic surgeons, cardiac anesthetists and imaging specialists is best suited to make decisions between open

of symptoms in the frailest of patients (Ussia *et al.*, 2010; Zahn *et al.*, 2011).

Hunaid A. Vohra, Robert N. Whistance and Sunil K. Ohri


## **Transcatheter Aortic Valve Implantation**

Hunaid A. Vohra, Robert N. Whistance and Sunil K. Ohri

*Wessex Cardiac Centre, Southampton University Hospitals Trust UK* 

## **1. Introduction**

220 Aortic Valve

Webb J, Altwegg L, Boone RH, Cheung A, Ye J, Lichenstein S, Lee M, Masson JB, Thompson

Wijesinghe N, Ye J, Rodés-Cabau J, Cheung A, Velianou JL, Natarajan MK, Dumont E,

*Circulation.* Vol. 119, pp. 3009-3016 ISSN 0009-7322

8798

C, Moss R, Carere R, Munt B, Nietlispach F & Humphries K. (2009). Transcatheter Aortic Valve Implantation: Impact On Clinical and Valve-Related Outcome.

Nietlispach F, Gurvitch R, Wood DA, Tay E, Webb JG. (2010). Journal American College Cardiology: Cardiovascular Interventions. Volume 3, 1122-1125 ISSN: 1936-

> Calcific aortic stenosis (AS) is the most common form of degenerative heart valve disease in developed countries. AS predominantly affects the elderly with a prevalence of 4% in people over the age of 80 years. With increasing longevity and an ageing population the absolute number of people afflicted with AS is set to rise. The clinical course of the disease is insidious at first (Ross & Braunwald, 1968), but is followed by rapid progression once symptoms of congestive cardiac failure, angina and syncope develop (Cheitlin *et al.*, 1979; Otto *et al.*, 1989; Davies *et al.*, 1991; Peter *et al.*, 1993). If left untreated mortality exceeds 50% at two years (Kelly *et al.*, 1988; Turina *et al.*, 1987) and AS is, therefore, set to become a major public health problem in the ensuing decades.

> Aortic valve replacement (AVR) has been the gold standard intervention for AS for more than 40 years (Charlson *et al.*, 2006) and over 60,000 procedures are performed annually in the European Union. However, one third of patients are denied access to surgery, often due to their advanced age (Iung *et al.*, 2003, 2005). Percutaneous balloon valvuloplasty was heralded initially as a promising breakthrough for high-risk symptomatic patients with AS (Cribier *et al.*, 1986). However, long-term follow-up has yielded unacceptable rates of restenosis and poor event-free survival (Otto *et al.*, 1994; Lieberman *et al.*, 1995). Balloon valvuloplasty is now recommended only as a bridge to emergency AVR or in the palliation of symptoms in the frailest of patients (Ussia *et al.*, 2010; Zahn *et al.*, 2011).

> Transcatheter aortic valve implantation (TAVI) has recently emerged as an effective therapeutic alternative to conventional AVR for high-risk AS patients. TAVI was developed initially in porcine models (Anderson *et al.*, 1992), but it took a decade for this technology to be translated to humans (Cribier *et al.*, 2002). Initially, an antegrade transseptal approach was used, but this has now been superseded by transapical (Ye *et al.*, 2006; Webb *et al.*, 2007) and retrograde percutaneous techniques (transarterial or transaxillary) (Webb *et al.*, 2006). The range of different approaches has increased the feasibility of TAVI in patients with relative contraindications, such as extensive peripheral vascular disease, porcelain aorta and thoracic radiotherapy. TAVI is also less invasive than open AVR and permits replacement of the native diseased valve in the beating heart without the need for sternotomy and cardiopulmonary bypass. Consequently, TAVI may be less influenced by a patient's comorbidities and may facilitate faster recovery.

> A multidisciplinary team consisting of interventional cardiologists, cardiothoracic surgeons, cardiac anesthetists and imaging specialists is best suited to make decisions between open

Transcatheter Aortic Valve Implantation 223

1992; Nashef *et al.*, 1999; Nowicki et al, 2004; O'Brien *et al.*, 2009; Parsonnet *et al.*, 1989; Pons *et al.*, 1997; Roques *et al.*, 1995; Roques *et al.*, 1999; Tremblay *et al.*, 1993; Tu *et al.*, 1995). These systems are usually derived from multivariate analyses of preoperative and operative variables believed to influence outcomes in large cohorts of cardiothoracic patients. Risk stratification tools may be either generic for all open cardiothoracic procedures or specific to heart-valve interventions. The focus of this subsection is to discuss some of the widely used

The Parsonnet score was one of the first mortality indicators developed to calculate average risk estimates of death in adult cardiothoracic patients (Parsonnet *et al.*, 1989). The model allocates "additive" points for 14 risk factors associated with perioperative mortality, which are subsequently used to assign a percentage probability of death. The factors include female gender, obesity, diabetes, hypertension, low ejection fraction, increased age, first or second reoperation, preoperative intra-aortic balloon pump, emergency presentation from the cardiac catheter laboratory, dialysis dependency, catastrophic clinical state, valve surgery and combined valve surgery and coronary artery bypass grafts (CABG). In addition to overestimating mortality in high-risk cases, the Parsonnet score has been criticized for being subjective and including some factors that are now known not to be linked to early postoperative death (Gabrielle *et al.*, 1997). The Parsonnet score also omits other potentially important factors, such as the urgency of surgery, which is widely considered to be strongly

The Parsonnet score has been superseded by the European System for Cardiac Operative Risk Evaluation (EuroSCORE). The EuroSCORE was developed by analyzing survival outcomes for 19,030 cardiac surgical patients in eight European countries (Roques *et al.*, 1999). Logistic regression was used to reduce a list of 97 potential postoperative mortality risk factors to 18 independent variables with odds ratios of >1 predictive of death. It bears many similarities to the Parsonnet system and involves calculation of percentage predicted mortality by addition of points ascribed to various risk factors to produce an "additive" mortality score (Table 1). For AVR, the EuroSCORE study reported an overall procedural mortality of 6% and mortality in the absence of risk factors of 1.1%. The precision and accuracy of the EuroSCORE has been demonstrated in numerous studies (Kobayashi et al, 2009; Wendt *et al.*, 2010), but it has several limitations that compromise its validity in sufferers of severe, symptomatic AS. First, the data is derived from patients undergoing cardiac surgery for heterogeneous indications, including a high percentage of cases of isolated coronary artery bypass grafts (63.9%). Risk factors are, therefore, not specific to AVR. Secondly, there is evidence that the EuroSCORE overestimates mortality in high-risk patients or is inaccurate in those with unusual combinations of risk factors (Brown *et al.*, 2008; Dewey *et al.*, 2008). This can exaggerate mortality estimates and may mean that patients are not offered AVR, even if they are potential surgical candidates. The "logistic" EuroSCORE has since been introduced for use in high-risk individuals because of its greater accuracy (Roques *et al.*, 2003). Nonetheless, it still suffers from problems with clinical relevance. Consequently, the "additive" and "logistic" algorithms have been combined to form the "modified" EuroSCORE (Nissinen *et al.*, 2009). Further problems with the EuroSCORE relate to the fact that it does not take into account local institutional outcomes (Vahanian *et al.*, 2008) or characteristics unique to certain AS patients (e.g. previous CABG with patent grafts, porcelain aorta or thoracic radiotherapy). These factors may confer additional risks that can alter the choice of intervention at the institutional level (Robes-

associated with perioperative outcome (Wynne-Jones *et al.*, 2000).

risk stratification tools.

Cabau *et al.*, 2010).

AVR, TAVI and medical management. This ensures apt patient selection and prompt delivery of care. There has been a rapid expansion in the number of studies investigating TAVI in the last five years and these have demonstrated promising results in terms of feasibility, safety and efficacy. Doubts remain, however, about the long-term durability of TAVI implants and their disposition to valvular dysfunction and about the need for reoperation. This chapter discusses the selection of patients for TAVI, techniques of implantation, clinical and patient-reported outcomes and future directions of research and development.

## **2. Patient selection**

Clinical decision-making in patients with AS is complex and from the outset requires a patient-centered approach and the involvement of a multidisciplinary team. Open AVR is associated with excellent clinical and functional outcomes in large modern series (Brown *et al.*, 2009; Malaisrie *et al.*, 2010) and, at present, is the gold standard intervention for patients with severe, symptomatic AS (Bonow *et al.*, 2006). However, patient selection is controversial. Central to the controversy is a belief that elderly patients, especially those with major comorbidities or with complications of AS (e.g. left ventricular dysfunction) present too great an operative risk or lack sufficient life expectancy to justify surgical AVR (Asimakopoulos *et al.*, 1997; Connolly *et al.*, 2000; Elayda *et al.*, 1993; Monin *et al.*, 2003). AVR is highly invasive and requires sternotomy, cardiopulmonary bypass, hypothermic cardioplegic arrest and cardiotomy. These factors expose patients to certain deleterious effects on end-organs, including ischemia, reperfusion injury, systematic inflammatory response, surgical trauma and oxidative stress (Anselmi *et al.*, 2004). In 2003, the Euro Heart Survey on Valvular Heart Disease reported that approximately one third of patients with severe, symptomatic AS are denied potentially life-saving surgery because of concerns over age, comorbidities and likely longevity (Iung *et al.*, 2003, 2005). This is in spite of some recent series demonstrating promising safe and efficacious outcomes for AVR in octogenarians (Melby *et al.*, 2007; Florath *et al.*, 2010; Leontyev *et al.*, 2009).

TAVI is a novel approach to AS in high-risk patients and provides an alternative management option that negates invasive surgery. TAVI is still in its infancy in terms of the operator learning curve, the level of technological development of the implants and the evidence base required to determine which patients are most likely to benefit. At present, TAVI is recommended only for patients not considered suitable for open AVR and who have a life expectancy of greater than one year (Vahanian *et al.*, 2008). This section will describe some of the available patient risk stratification tools used for clinical decisionmaking in AS, the current indications and contraindications to TAVI, and possible diversification in the use of minimally invasive aortic valve interventions.

### **2.1 Risk stratification tools**

The choice between treatment options in AS should ideally be based upon a shared decision between the fully informed patient and a multidisciplinary team who guide the patient to the required information. Part of this process includes an assessment by physicians of the likely risk of mortality and serious morbidity of undergoing open AVR. A range of scoring systems have been developed that are designed to assist surgeons in such risk stratification (Ambler *et al.*, 2005; Edwards *et al.*, 2001; Florath *et al.*, 2003; Hattler *et al.*, 1994; Higgins *et al.*,

AVR, TAVI and medical management. This ensures apt patient selection and prompt delivery of care. There has been a rapid expansion in the number of studies investigating TAVI in the last five years and these have demonstrated promising results in terms of feasibility, safety and efficacy. Doubts remain, however, about the long-term durability of TAVI implants and their disposition to valvular dysfunction and about the need for reoperation. This chapter discusses the selection of patients for TAVI, techniques of implantation, clinical and patient-reported outcomes and future directions of research and

Clinical decision-making in patients with AS is complex and from the outset requires a patient-centered approach and the involvement of a multidisciplinary team. Open AVR is associated with excellent clinical and functional outcomes in large modern series (Brown *et al.*, 2009; Malaisrie *et al.*, 2010) and, at present, is the gold standard intervention for patients with severe, symptomatic AS (Bonow *et al.*, 2006). However, patient selection is controversial. Central to the controversy is a belief that elderly patients, especially those with major comorbidities or with complications of AS (e.g. left ventricular dysfunction) present too great an operative risk or lack sufficient life expectancy to justify surgical AVR (Asimakopoulos *et al.*, 1997; Connolly *et al.*, 2000; Elayda *et al.*, 1993; Monin *et al.*, 2003). AVR is highly invasive and requires sternotomy, cardiopulmonary bypass, hypothermic cardioplegic arrest and cardiotomy. These factors expose patients to certain deleterious effects on end-organs, including ischemia, reperfusion injury, systematic inflammatory response, surgical trauma and oxidative stress (Anselmi *et al.*, 2004). In 2003, the Euro Heart Survey on Valvular Heart Disease reported that approximately one third of patients with severe, symptomatic AS are denied potentially life-saving surgery because of concerns over age, comorbidities and likely longevity (Iung *et al.*, 2003, 2005). This is in spite of some recent series demonstrating promising safe and efficacious outcomes for AVR in octogenarians

TAVI is a novel approach to AS in high-risk patients and provides an alternative management option that negates invasive surgery. TAVI is still in its infancy in terms of the operator learning curve, the level of technological development of the implants and the evidence base required to determine which patients are most likely to benefit. At present, TAVI is recommended only for patients not considered suitable for open AVR and who have a life expectancy of greater than one year (Vahanian *et al.*, 2008). This section will describe some of the available patient risk stratification tools used for clinical decisionmaking in AS, the current indications and contraindications to TAVI, and possible

The choice between treatment options in AS should ideally be based upon a shared decision between the fully informed patient and a multidisciplinary team who guide the patient to the required information. Part of this process includes an assessment by physicians of the likely risk of mortality and serious morbidity of undergoing open AVR. A range of scoring systems have been developed that are designed to assist surgeons in such risk stratification (Ambler *et al.*, 2005; Edwards *et al.*, 2001; Florath *et al.*, 2003; Hattler *et al.*, 1994; Higgins *et al.*,

(Melby *et al.*, 2007; Florath *et al.*, 2010; Leontyev *et al.*, 2009).

diversification in the use of minimally invasive aortic valve interventions.

development.

**2. Patient selection** 

**2.1 Risk stratification tools** 

1992; Nashef *et al.*, 1999; Nowicki et al, 2004; O'Brien *et al.*, 2009; Parsonnet *et al.*, 1989; Pons *et al.*, 1997; Roques *et al.*, 1995; Roques *et al.*, 1999; Tremblay *et al.*, 1993; Tu *et al.*, 1995). These systems are usually derived from multivariate analyses of preoperative and operative variables believed to influence outcomes in large cohorts of cardiothoracic patients. Risk stratification tools may be either generic for all open cardiothoracic procedures or specific to heart-valve interventions. The focus of this subsection is to discuss some of the widely used risk stratification tools.

The Parsonnet score was one of the first mortality indicators developed to calculate average risk estimates of death in adult cardiothoracic patients (Parsonnet *et al.*, 1989). The model allocates "additive" points for 14 risk factors associated with perioperative mortality, which are subsequently used to assign a percentage probability of death. The factors include female gender, obesity, diabetes, hypertension, low ejection fraction, increased age, first or second reoperation, preoperative intra-aortic balloon pump, emergency presentation from the cardiac catheter laboratory, dialysis dependency, catastrophic clinical state, valve surgery and combined valve surgery and coronary artery bypass grafts (CABG). In addition to overestimating mortality in high-risk cases, the Parsonnet score has been criticized for being subjective and including some factors that are now known not to be linked to early postoperative death (Gabrielle *et al.*, 1997). The Parsonnet score also omits other potentially important factors, such as the urgency of surgery, which is widely considered to be strongly associated with perioperative outcome (Wynne-Jones *et al.*, 2000).

The Parsonnet score has been superseded by the European System for Cardiac Operative Risk Evaluation (EuroSCORE). The EuroSCORE was developed by analyzing survival outcomes for 19,030 cardiac surgical patients in eight European countries (Roques *et al.*, 1999). Logistic regression was used to reduce a list of 97 potential postoperative mortality risk factors to 18 independent variables with odds ratios of >1 predictive of death. It bears many similarities to the Parsonnet system and involves calculation of percentage predicted mortality by addition of points ascribed to various risk factors to produce an "additive" mortality score (Table 1). For AVR, the EuroSCORE study reported an overall procedural mortality of 6% and mortality in the absence of risk factors of 1.1%. The precision and accuracy of the EuroSCORE has been demonstrated in numerous studies (Kobayashi et al, 2009; Wendt *et al.*, 2010), but it has several limitations that compromise its validity in sufferers of severe, symptomatic AS. First, the data is derived from patients undergoing cardiac surgery for heterogeneous indications, including a high percentage of cases of isolated coronary artery bypass grafts (63.9%). Risk factors are, therefore, not specific to AVR. Secondly, there is evidence that the EuroSCORE overestimates mortality in high-risk patients or is inaccurate in those with unusual combinations of risk factors (Brown *et al.*, 2008; Dewey *et al.*, 2008). This can exaggerate mortality estimates and may mean that patients are not offered AVR, even if they are potential surgical candidates. The "logistic" EuroSCORE has since been introduced for use in high-risk individuals because of its greater accuracy (Roques *et al.*, 2003). Nonetheless, it still suffers from problems with clinical relevance. Consequently, the "additive" and "logistic" algorithms have been combined to form the "modified" EuroSCORE (Nissinen *et al.*, 2009). Further problems with the EuroSCORE relate to the fact that it does not take into account local institutional outcomes (Vahanian *et al.*, 2008) or characteristics unique to certain AS patients (e.g. previous CABG with patent grafts, porcelain aorta or thoracic radiotherapy). These factors may confer additional risks that can alter the choice of intervention at the institutional level (Robes-Cabau *et al.*, 2010).

Transcatheter Aortic Valve Implantation 225

Association of Cardiothoracic Surgery (EACTS), the European Society of Cardiology (ECS), and the European Association of Percutaneous Cardiovascular Intervention (EAPCI) issued a joint statement describing a four-stage assessment procedure to be undertaken to differentiate between treatment options in AS and to confirm suitability for TAVI (Vahanian *et al.*, 2008). The first stage involves echocardiography to confirm the diagnosis and severity of AS, and to exclude significant aortic regurgitation. TAVI is presently restricted to cases of severe AS because of the proven effectiveness of AVR. The inclusion criteria of welldesigned trials offer some insight into the definition of the severity of AS at echocardiography. In THE PARTNER TRIAL: Placement of AoRTic TraNscathetER Valve Trial, patients were randomized to interventions (AVR, TAVI, or medical management) only if proved to have an aortic valve area (AVA) of < 0.8 cm2 (or an indexed AVA of < 0.6 cm2) or a peak aortic valve gradient of > 40 mmHg (or a peak velocity of > 4.0 m/s) (Leon *et al.*, 2010). Analogous values should be used when deciding whether TAVI is appropriate for particular individuals, and many studies have used similar cut-offs to assign patients to treatment with TAVI (Clavel *et al.*, 2010; Malaisrie *et al.*, 2011; Zahn *et al.*, 2011). In certain circumstances, low-dose dobutamine echocardiography may also be of value to distinguish between severe AS and the rare "pseudo-severe" AS, especially in patients with a low left ventricular (LV) ejection fraction and a low aortic transvalvular gradient (Bonow *et al.*, 2006;

The second stage in TAVI assessment involves an accurate evaluation of the presenting symptoms and clinical findings. Current guidelines recommend that TAVI should be undertaken only when symptoms are directly attributable to AS (Vahanian *et al.*, 2008). Several concurrent diseases may mimic the symptom profile of AS, such as chronic obstructive pulmonary disease or pulmonary hypertension. A clear history and chronology of the symptoms of dyspnea, chest pain, and syncope are, therefore, required, and when the diagnosis is uncertain, biomarkers of increased myocardial mechanical load, such as betatype natriuretic peptide (BNP), may be of value (Vahanian *et al.*, 2010). Clinical confirmation of the diagnosis is necessary because of concerns over the long-term durability of TAVI

Arguably, the most complex stage when assessing patients with AS is deciding between AVR, TAVI, and conservative management. This requires multidisciplinary team evaluation of the possible risks of open AVR and TAVI, predicted life expectancy, and quality of life. To evaluate the risk of operative mortality and serious morbidity, it is recommended that clinical judgment be combined with the scores from two risk stratification tools (Vahanian *et al.*, 2008). This combination allows for an objective risk assessment, while ensuring that additional factors not included in risk stratification tools (e.g. porcelain aorta, previous CABG with patent grafts, previous thoracic radiotherapy, or liver cirrhosis) are taken into account. Typically, mortality > 20% calculated using the logistic EuroSCORE or > 10% using the STS-PROM are seen as high-risk indicators that would preclude open AVR in most cases (Leon *et al.*, 2010; Vahanian *et al.*, 2008; Vahanian *et al.*, 2010). However, scoring systems should not necessarily be viewed as a substitute for experienced clinical judgment or informed patient choice. As part of the multidisciplinary approach, it is recommended that patient-reported outcomes, such as health-related quality of life (HRQL), be considered. The use of validated HRQL tools should be combined with clinical parameters to assist in the

The final stage before TAVI insertion in high-risk patients with severe, symptomatic AS includes an assessment of the feasibility and contraindications to TAVI. The first-line

shared decision-making process (Lee *et al.*, 2006; Vahanian *et al.*, 2008).

Vahanian *et al.*, 2007).

devices.


Table 1. Risk factors for mortality in patients undergoing cardiac surgery based on the EuroSCORE dataset (reproduced from Roques *et al.*, 1999; LVEF = left ventricular ejection fraction)

Recently, emphasis has been placed on developing risk stratification tools with greater reliability and validity in patients with valvular heart disease. The Ambler score was published in 2005 after it had been field-tested in 32,839 consecutive patients undergoing heart valve surgery in the UK (Ambler *et al.*, 2005). It identified 14 risk factors associated with in-hospital mortality (operative priority, age, renal failure, operation sequence, ejection fraction, concomitant tricuspid valve surgery, type of valve surgery, concomitant CABG, body mass index, preoperative arrhythmia, diabetes, gender, and hypertension). Its development involved robust methodology in a large cohort of patients, but it has not been widely adopted, perhaps because of concerns over external validity in non-UK populations. The Society of Thoracic Surgeons has since built on the advantages of the Ambler score by producing a tool that assists in predicting nine different postoperative variables (STS-PROM; O'Brien *et al.*, 2009). These are mortality, permanent stroke, renal failure, prolonged ventilation, deep sternal wound infection, reoperation, a composite endpoint of mortality and major morbidity, and short and prolonged postoperative stay. In addition to extending predictive models to several causes of major morbidity, the STS-PROM has been shown to be more reliable than the EuroSCORE in estimating mortality in high-risk patients (Dewey *et al.*, 2008). At present, most major studies involving the selection of patients for TAVI use the EuroSCORE and STS-PROM.

#### **2.2 Indications and assessment of suitability for TAVI**

TAVI is recommended for use only in patients with calcified pure or predominant AS, and not in cases where aortic regurgitation is the primary pathology. In 2008, the European

Recently, emphasis has been placed on developing risk stratification tools with greater reliability and validity in patients with valvular heart disease. The Ambler score was published in 2005 after it had been field-tested in 32,839 consecutive patients undergoing heart valve surgery in the UK (Ambler *et al.*, 2005). It identified 14 risk factors associated with in-hospital mortality (operative priority, age, renal failure, operation sequence, ejection fraction, concomitant tricuspid valve surgery, type of valve surgery, concomitant CABG, body mass index, preoperative arrhythmia, diabetes, gender, and hypertension). Its development involved robust methodology in a large cohort of patients, but it has not been widely adopted, perhaps because of concerns over external validity in non-UK populations. The Society of Thoracic Surgeons has since built on the advantages of the Ambler score by producing a tool that assists in predicting nine different postoperative variables (STS-PROM; O'Brien *et al.*, 2009). These are mortality, permanent stroke, renal failure, prolonged ventilation, deep sternal wound infection, reoperation, a composite endpoint of mortality and major morbidity, and short and prolonged postoperative stay. In addition to extending predictive models to several causes of major morbidity, the STS-PROM has been shown to be more reliable than the EuroSCORE in estimating mortality in high-risk patients (Dewey *et al.*, 2008). At present, most major studies involving the selection of patients for TAVI use the

TAVI is recommended for use only in patients with calcified pure or predominant AS, and not in cases where aortic regurgitation is the primary pathology. In 2008, the European

Age per 5 years after 60 years 1.1 ± 0.007 0.001 1 Female gender 1.4 ± 0.128 0.001 1 Preoperative creatinine > 200 μmol/L 1.9 ± 0.256 0.001 2 Extracardiac arteriopathy 1.9 ± 0.376 0.001 2 Pulmonary disease 1.6 ± 0.284 0.006 1 Neurological dysfunction 2.3 ± 0.584 0.001 2 Previous cardiac surgery 2.6 ± 0.324 0.001 3 Recent myocardial infarction 1.6 ± 0.208 0.001 2 LVEF 30-50% 1.5 ± 0.138 0.001 1 LVEF < 30% 2.5 ± 0.340 0.001 3 Pulmonary hypertension (> 60 mmHg) 2.0 ± 0.423 0.001 2 Active endocarditis 2.5 ± 0.678 0.001 3 Unstable angina 1.5 ± 0.202 0.001 2 Emergency operation 2.8 ± 0.440 0.001 2 Critical perioperative condition 2.2 ± 0.319 0.001 3 Ventricular septal rupture 3.8 ± 1.735 0.002 4 Noncoronary surgery 1.6 ± 0.170 0.001 2 Thoracic aortic surgery 3.2 ± 0.650 0.001 3 Table 1. Risk factors for mortality in patients undergoing cardiac surgery based on the EuroSCORE dataset (reproduced from Roques *et al.*, 1999; LVEF = left ventricular ejection

**Standard error**

**P-value Additional %** 

**mortality** 

**Variable Odds ratio ±** 

fraction)

EuroSCORE and STS-PROM.

**2.2 Indications and assessment of suitability for TAVI** 

Association of Cardiothoracic Surgery (EACTS), the European Society of Cardiology (ECS), and the European Association of Percutaneous Cardiovascular Intervention (EAPCI) issued a joint statement describing a four-stage assessment procedure to be undertaken to differentiate between treatment options in AS and to confirm suitability for TAVI (Vahanian *et al.*, 2008). The first stage involves echocardiography to confirm the diagnosis and severity of AS, and to exclude significant aortic regurgitation. TAVI is presently restricted to cases of severe AS because of the proven effectiveness of AVR. The inclusion criteria of welldesigned trials offer some insight into the definition of the severity of AS at echocardiography. In THE PARTNER TRIAL: Placement of AoRTic TraNscathetER Valve Trial, patients were randomized to interventions (AVR, TAVI, or medical management) only if proved to have an aortic valve area (AVA) of < 0.8 cm2 (or an indexed AVA of < 0.6 cm2) or a peak aortic valve gradient of > 40 mmHg (or a peak velocity of > 4.0 m/s) (Leon *et al.*, 2010). Analogous values should be used when deciding whether TAVI is appropriate for particular individuals, and many studies have used similar cut-offs to assign patients to treatment with TAVI (Clavel *et al.*, 2010; Malaisrie *et al.*, 2011; Zahn *et al.*, 2011). In certain circumstances, low-dose dobutamine echocardiography may also be of value to distinguish between severe AS and the rare "pseudo-severe" AS, especially in patients with a low left ventricular (LV) ejection fraction and a low aortic transvalvular gradient (Bonow *et al.*, 2006; Vahanian *et al.*, 2007).

The second stage in TAVI assessment involves an accurate evaluation of the presenting symptoms and clinical findings. Current guidelines recommend that TAVI should be undertaken only when symptoms are directly attributable to AS (Vahanian *et al.*, 2008). Several concurrent diseases may mimic the symptom profile of AS, such as chronic obstructive pulmonary disease or pulmonary hypertension. A clear history and chronology of the symptoms of dyspnea, chest pain, and syncope are, therefore, required, and when the diagnosis is uncertain, biomarkers of increased myocardial mechanical load, such as betatype natriuretic peptide (BNP), may be of value (Vahanian *et al.*, 2010). Clinical confirmation of the diagnosis is necessary because of concerns over the long-term durability of TAVI devices.

Arguably, the most complex stage when assessing patients with AS is deciding between AVR, TAVI, and conservative management. This requires multidisciplinary team evaluation of the possible risks of open AVR and TAVI, predicted life expectancy, and quality of life. To evaluate the risk of operative mortality and serious morbidity, it is recommended that clinical judgment be combined with the scores from two risk stratification tools (Vahanian *et al.*, 2008). This combination allows for an objective risk assessment, while ensuring that additional factors not included in risk stratification tools (e.g. porcelain aorta, previous CABG with patent grafts, previous thoracic radiotherapy, or liver cirrhosis) are taken into account. Typically, mortality > 20% calculated using the logistic EuroSCORE or > 10% using the STS-PROM are seen as high-risk indicators that would preclude open AVR in most cases (Leon *et al.*, 2010; Vahanian *et al.*, 2008; Vahanian *et al.*, 2010). However, scoring systems should not necessarily be viewed as a substitute for experienced clinical judgment or informed patient choice. As part of the multidisciplinary approach, it is recommended that patient-reported outcomes, such as health-related quality of life (HRQL), be considered. The use of validated HRQL tools should be combined with clinical parameters to assist in the shared decision-making process (Lee *et al.*, 2006; Vahanian *et al.*, 2008).

The final stage before TAVI insertion in high-risk patients with severe, symptomatic AS includes an assessment of the feasibility and contraindications to TAVI. The first-line

Transcatheter Aortic Valve Implantation 227

• Inability to access the apex of the left ventricle due to anatomical constraints (e.g. chest

Implants used in open AVR are either mechanical or bioprosthetic. Mechanical valves have the advantage of long-term durability, but require life-long anticoagulation with the associated risk of major hemorrhage. Patients fitted with bioprosthetic valves do not need to take anticoagulants, but the chance of valvular degeneration increases with time. Mechanical valves are, therefore, usually restricted to younger patients, while bioprosthetic implants are used more frequently in the elderly population, in which the chance of surviving to revision surgery is low. Nonetheless, valvular degeneration occurs in a proportion of patients, and treatment typically necessitates revision AVR, which is

In 2008 the first case of valve-in-valve (VIV) TAVI was reported in an 82-year-old patient with valvular degeneration of a Carpentier-Edwards Perimount aortic valve (Walther *et al.*, 2008a). This pioneering procedure was performed off-pump using a transapical approach. VIV is an attractive technique that involves placing the implant within the previous prosthetic valve, abutting the degenerated valve leaflets up against the aortic annulus. Typically, transapical access is preferred for VIV procedures, although reports have recently emerged describing the transaxillary (Sharp *et al.*, 2010) and trans-subclavian approaches (Olsen *et al.*, 2010). The Edwards SAPIEN valve may be better suited to VIV implantation (Kempfert *et al.*, 2010), although the Medtronic CoreValve system has been successfully used

There are promising early results for VIV, although reports tend to be anecdotal or restricted to small case series, and no large comparative studies are currently available (Ferrari *et al.*, 2010; Kempfert *et al.*, 2010; Khawaja *et al.*, 2010; Olsen *et al.*, 2010; Sharp *et al.*, 2010). Transvalvular gradients post-procedure are usually satisfactory (Ferrari *et al.*, 2010; Kempfert *et al.*, 2010; Walther *et al.*, 2008a), while residual aortic regurgitation tends to be minimal (Kempfert *et al.*, 2010). Severe paravalvular leak is a feared complication of TAVI, but several authors have described its successful treatment with rescue VIV (Rodes-Cabau *et al.*, 2009; Taramasso *et al.*, 2010). Concerns exist over excessive transvalvular gradients of VIV implants in patients fitted previously with small-diameter prostheses (e.g. < 23 mm in diameter) (Ferrari *et al.*, 2010). For TAVI, the diameter of the device is typically oversized by 10-20% in relation to the aortic annulus, but for VIV undersizing is preferred, which is currently not feasible in some patients with small annuli because of the size of available

implants. At present, VIV is off-label in many countries, including the United States.

TAVI is minimally invasive and has consistently demonstrated promising outcomes in highrisk patients (see Section 4). In the future, it is likely that the indications for TAVI will be expanded to include younger patients at low operative risk. However, caution is advised for

• Transverse ascending arch (for balloon expandable devices).

• Previous surgical patch of the left ventricle (e.g. Dor procedure).

• Aortic aneurysm with extensive mural thrombus.

For the transapical approach, contraindications are:

• Severely impaired respiratory function.

• Coarctation of the aorta.

• Calcified pericardium.

**2.4 Expanding the role of TAVI** 

deformity).

inherently high-risk.

(Khawaja *et al.*, 2010).

investigation is coronary angiography to identify occlusive lesions in need of revascularization. Lesions amenable to percutaneous angioplasty and stenting can be treated either before, during, or after TAVI. The decision on the timing of revascularization is complex and should be tailored to individual cases. Patients with proximal coronary stenoses may not be appropriate candidates for TAVI (Vahanian *et al.*, 2008) because of concerns over occlusion of the coronary ostia by the device (Gogas *et al.*, 2011; Gurvitch *et al.*, 2011). If coronary stenoses can be managed only surgically, then a choice must be made between high-risk AVR and the poor outcomes associated with medical management and balloon valvuloplasty.

Determining the diameter of the aortic annulus is a prerequisite to TAVI operational planning and ensures that an appropriately sized implant is deployed. This can be accomplished using either invasive techniques (e.g. aortography as part of balloon valvuloplasty) or noninvasive imaging modalities (e.g. echocardiography, multislice high resolution computed tomography, or magnetic resonance imaging) (Tops *et al.*, 2008; Vahanian *et al.*, 2008). Transthoracic echocardiography has been shown to underestimate the size of the aortic annulus, and should be supplemented with transesophageal echocardiography when borderline sizes result in doubt over the feasibility of the procedure (Moss *et al.*, 2008). Accurate sizing prior to TAVI is necessary to prevent paravalvular leak and rupture of the aortic annulus. The peripheral vasculature must also be imaged, in particular the aortic arch, descending aorta, and iliac vessels. This can be achieved with either formal or computed tomography angiography. Gadolinium magnetic resonance angiography is an alternative in patients with impaired renal function. Both the size and tortuosity of the vessels are important because they affect access and help to decide between the transarterial and transapical approaches.

## **2.3 Contraindications to TAVI**

For technical reasons, TAVI is not possible in all high-risk patients with AS, but using different access ports has increased the number of patients who can be successfully treated. General contraindications to TAVI include:


There are a number of contraindications specific to the type of approach. For the transfemoral approach, these are:


investigation is coronary angiography to identify occlusive lesions in need of revascularization. Lesions amenable to percutaneous angioplasty and stenting can be treated either before, during, or after TAVI. The decision on the timing of revascularization is complex and should be tailored to individual cases. Patients with proximal coronary stenoses may not be appropriate candidates for TAVI (Vahanian *et al.*, 2008) because of concerns over occlusion of the coronary ostia by the device (Gogas *et al.*, 2011; Gurvitch *et al.*, 2011). If coronary stenoses can be managed only surgically, then a choice must be made between high-risk AVR and the poor outcomes associated with medical management and

Determining the diameter of the aortic annulus is a prerequisite to TAVI operational planning and ensures that an appropriately sized implant is deployed. This can be accomplished using either invasive techniques (e.g. aortography as part of balloon valvuloplasty) or noninvasive imaging modalities (e.g. echocardiography, multislice high resolution computed tomography, or magnetic resonance imaging) (Tops *et al.*, 2008; Vahanian *et al.*, 2008). Transthoracic echocardiography has been shown to underestimate the size of the aortic annulus, and should be supplemented with transesophageal echocardiography when borderline sizes result in doubt over the feasibility of the procedure (Moss *et al.*, 2008). Accurate sizing prior to TAVI is necessary to prevent paravalvular leak and rupture of the aortic annulus. The peripheral vasculature must also be imaged, in particular the aortic arch, descending aorta, and iliac vessels. This can be achieved with either formal or computed tomography angiography. Gadolinium magnetic resonance angiography is an alternative in patients with impaired renal function. Both the size and tortuosity of the vessels are important because they affect access and help to decide between

For technical reasons, TAVI is not possible in all high-risk patients with AS, but using different access ports has increased the number of patients who can be successfully treated.

• Aortic annulus diameter < 18 mm or > 25 mm for balloon expandable implants, and <

• Bicuspid aortic valves that may lead to incomplete deployment of the device and

• Heavy asymmetrical aortic valve calcification because of concerns over occlusion of the

• Aortic root diameter of > 45 mm at the aorto-tubular junction for self-expandable

There are a number of contraindications specific to the type of approach. For the

• Iliac artery diameter of < 6 mm to < 9 mm, depending on the type of device used.

• Low position of the coronary ostia (< 8 mm from the aortic annulus).

balloon valvuloplasty.

the transarterial and transapical approaches.

General contraindications to TAVI include:

paravalvular leak (Zegdi *et al.*, 2008).

coronary ostia (Webb *et al.*, 2006).

• Severe organic mitral regurgitation.

• Severely calcified or tortuous iliac arteries.

• Severely angulated aorta or atherosclerotic aortic arch.

• Previous aorto-femoral bypass grafts.

transfemoral approach, these are:

20 mm or > 27 mm for self-expandable devices.

**2.3 Contraindications to TAVI** 

devices.

• LV thrombus.


#### **2.4 Expanding the role of TAVI**

Implants used in open AVR are either mechanical or bioprosthetic. Mechanical valves have the advantage of long-term durability, but require life-long anticoagulation with the associated risk of major hemorrhage. Patients fitted with bioprosthetic valves do not need to take anticoagulants, but the chance of valvular degeneration increases with time. Mechanical valves are, therefore, usually restricted to younger patients, while bioprosthetic implants are used more frequently in the elderly population, in which the chance of surviving to revision surgery is low. Nonetheless, valvular degeneration occurs in a proportion of patients, and treatment typically necessitates revision AVR, which is inherently high-risk.

In 2008 the first case of valve-in-valve (VIV) TAVI was reported in an 82-year-old patient with valvular degeneration of a Carpentier-Edwards Perimount aortic valve (Walther *et al.*, 2008a). This pioneering procedure was performed off-pump using a transapical approach. VIV is an attractive technique that involves placing the implant within the previous prosthetic valve, abutting the degenerated valve leaflets up against the aortic annulus. Typically, transapical access is preferred for VIV procedures, although reports have recently emerged describing the transaxillary (Sharp *et al.*, 2010) and trans-subclavian approaches (Olsen *et al.*, 2010). The Edwards SAPIEN valve may be better suited to VIV implantation (Kempfert *et al.*, 2010), although the Medtronic CoreValve system has been successfully used (Khawaja *et al.*, 2010).

There are promising early results for VIV, although reports tend to be anecdotal or restricted to small case series, and no large comparative studies are currently available (Ferrari *et al.*, 2010; Kempfert *et al.*, 2010; Khawaja *et al.*, 2010; Olsen *et al.*, 2010; Sharp *et al.*, 2010). Transvalvular gradients post-procedure are usually satisfactory (Ferrari *et al.*, 2010; Kempfert *et al.*, 2010; Walther *et al.*, 2008a), while residual aortic regurgitation tends to be minimal (Kempfert *et al.*, 2010). Severe paravalvular leak is a feared complication of TAVI, but several authors have described its successful treatment with rescue VIV (Rodes-Cabau *et al.*, 2009; Taramasso *et al.*, 2010). Concerns exist over excessive transvalvular gradients of VIV implants in patients fitted previously with small-diameter prostheses (e.g. < 23 mm in diameter) (Ferrari *et al.*, 2010). For TAVI, the diameter of the device is typically oversized by 10-20% in relation to the aortic annulus, but for VIV undersizing is preferred, which is currently not feasible in some patients with small annuli because of the size of available implants. At present, VIV is off-label in many countries, including the United States.

TAVI is minimally invasive and has consistently demonstrated promising outcomes in highrisk patients (see Section 4). In the future, it is likely that the indications for TAVI will be expanded to include younger patients at low operative risk. However, caution is advised for

Transcatheter Aortic Valve Implantation 229

Fig. 1. The Edwards SAPIEN Transcatheter aortic valve (Photo courtesy of Edwards

Fig. 2. Transcatheter aortic valve crimped onto the introducing catheter (Photo courtesy of

retrieval and greater accuracy when deployed. Examples include the DirectFlow (Direct Medical Flow Inc., Santa Rosa, CA, USA), Lotus (Boston Scientific Inc., Natick, MA, USA) and HLT (Heart Leaflet Technologies Inc., Maple Grove, MN, USA). Other systems incorporate features that allow fixation in the supracoronary aorta (Accurate, Symentis Inc., Lausanne, Switzerland; St Jude, St Jude Medical Inc., St. Paul, MN, USA) and anatomical guides to demonstrate the position of the native valve and coronary arteries, thus facilitating deployment (Engager, Medtronic Inc., USA; JenaClip, JenaValve Inc., Munich, Germany). At present, little is known about these novel valves in terms of their efficacy, safety, feasibility, and long-term durability. Further work is required to characterize their outcomes before

Initial reports of TAVI used a transseptal approach with access via the venous system (Cribier *et al.*, 2002). This route was technically challenging and not reproducible. Retrograde arterial approaches are now much more widely used. Access is gained typically through the femoral

they can be recommended for use in routine clinical practice.

Lifesciences)

Edwards Lifesciences)

**3.2 Percutaneous access** 

several reasons. Despite the lack of randomized clinical trial (RCT) data to support the use of open surgery, AVR has shown excellent long-term clinical, hemodynamic and functional outcomes in low-risk AS patients (Hammermeister *et al.*, 1993; Myken *et al.*, 1995; Peterseim *et al.*, 1999). Modern RCTs comparing AVR to medical management would be unethical because of a lack of clinical equipoise, leading to appropriate patients being denied access to a treatment with proven long-term effectiveness. The durability of TAVI implants is still a concern due to the technique's relative age. More work to characterize the long-term outcomes of TAVI is necessary before it is offered routinely to low-risk patients.

## **3. Devices and techniques**

The technology that underpins TAVI has evolved dramatically in the last two decades since its inception in animal models (Anderson *et al.*, 1992) and later realization in humans (Cribier *et al.*, 2002). At present, there are two major producers of TAVI devices that are used routinely in clinical practice (Edwards Lifesciences, Irvine, CA, USA & Medtronic Inc., Minneapolis, MN, USA). Technological development of both implants is on-going, and each has its advantages and disadvantages in different clinical situations. There are a variety of access sites through which TAVI can be performed, including directly through the left ventricular apex as well as the femoral, axillary and subclavian arteries. This section will describe the features of current and future implants and the techniques required for their implantation.

#### **3.1 TAVI devices**

The prototypic Cribier-Edwards TAVI device was one of the earliest deployed in humans. It has now been replaced by newer Edwards Lifesciences designs, including the SAPIEN transcatheter heart valve (THV) (Figure 1) and, more recently, by the SAPIEN XT. The SAPIEN devices are constructed from bovine pericardial leaflets mounted on a cobaltchromium frame. The device is balloon expandable and manufactured in 23-mm and 26-mm sizes, which allows it to be used in patients with aortic annulus diameters of between 18 mm and 25 mm. As the implant is deployed (Figure 2), it fixes within the aortic annulus without the need for stabilization in the ascending aorta (Webb and Cribier, 2010). Early devices required large-caliber delivery systems; however, the latest versions can be deployed through vessels with a minimum diameter of 6 mm. Transapical and percutaneous approaches can both be used with the SAPIEN system.

The Medtronic CoreValve consists of a porcine pericardial valve mounted on a nitinol selfexpandable metal frame. It is considerably longer than SAPIEN devices (53-55 mm versus 15-17 mm) and anchors distally in both the ascending aorta and supracoronary region. The CoreValve can be deployed through an access channel with a minimum diameter of 6 mm and can be used in patients with an aortic annulus diameter of between 20 mm and 27 mm. The device is not licensed for transapical use, but when used via a peripheral access vessel, it is associated with greater hemodynamic stability during deployment, more forgiving positioning and can be retrieved if sited incorrectly (Webb and Cribier, 2010). Despite this, the CoreValve suffers from a high incidence of post-procedural heart block, which requires prolonged cardiac monitoring, and pacemaker insertion is necessary in up to 40% of cases.

A number of next generation TAVI devices are currently undergoing clinical testing (Falk *et al.*, 2009; Low *et al.*, 2008; Schofer at al., 2008; Treede *et al.*, 2010). These are based generally on the self-expandable CoreValve system and allow for smaller caliber delivery systems,

several reasons. Despite the lack of randomized clinical trial (RCT) data to support the use of open surgery, AVR has shown excellent long-term clinical, hemodynamic and functional outcomes in low-risk AS patients (Hammermeister *et al.*, 1993; Myken *et al.*, 1995; Peterseim *et al.*, 1999). Modern RCTs comparing AVR to medical management would be unethical because of a lack of clinical equipoise, leading to appropriate patients being denied access to a treatment with proven long-term effectiveness. The durability of TAVI implants is still a concern due to the technique's relative age. More work to characterize the long-term

The technology that underpins TAVI has evolved dramatically in the last two decades since its inception in animal models (Anderson *et al.*, 1992) and later realization in humans (Cribier *et al.*, 2002). At present, there are two major producers of TAVI devices that are used routinely in clinical practice (Edwards Lifesciences, Irvine, CA, USA & Medtronic Inc., Minneapolis, MN, USA). Technological development of both implants is on-going, and each has its advantages and disadvantages in different clinical situations. There are a variety of access sites through which TAVI can be performed, including directly through the left ventricular apex as well as the femoral, axillary and subclavian arteries. This section will describe the features of current and future implants and the techniques required for their

The prototypic Cribier-Edwards TAVI device was one of the earliest deployed in humans. It has now been replaced by newer Edwards Lifesciences designs, including the SAPIEN transcatheter heart valve (THV) (Figure 1) and, more recently, by the SAPIEN XT. The SAPIEN devices are constructed from bovine pericardial leaflets mounted on a cobaltchromium frame. The device is balloon expandable and manufactured in 23-mm and 26-mm sizes, which allows it to be used in patients with aortic annulus diameters of between 18 mm and 25 mm. As the implant is deployed (Figure 2), it fixes within the aortic annulus without the need for stabilization in the ascending aorta (Webb and Cribier, 2010). Early devices required large-caliber delivery systems; however, the latest versions can be deployed through vessels with a minimum diameter of 6 mm. Transapical and percutaneous

The Medtronic CoreValve consists of a porcine pericardial valve mounted on a nitinol selfexpandable metal frame. It is considerably longer than SAPIEN devices (53-55 mm versus 15-17 mm) and anchors distally in both the ascending aorta and supracoronary region. The CoreValve can be deployed through an access channel with a minimum diameter of 6 mm and can be used in patients with an aortic annulus diameter of between 20 mm and 27 mm. The device is not licensed for transapical use, but when used via a peripheral access vessel, it is associated with greater hemodynamic stability during deployment, more forgiving positioning and can be retrieved if sited incorrectly (Webb and Cribier, 2010). Despite this, the CoreValve suffers from a high incidence of post-procedural heart block, which requires prolonged cardiac monitoring, and pacemaker insertion is necessary in up to 40% of cases. A number of next generation TAVI devices are currently undergoing clinical testing (Falk *et al.*, 2009; Low *et al.*, 2008; Schofer at al., 2008; Treede *et al.*, 2010). These are based generally on the self-expandable CoreValve system and allow for smaller caliber delivery systems,

approaches can both be used with the SAPIEN system.

outcomes of TAVI is necessary before it is offered routinely to low-risk patients.

**3. Devices and techniques** 

implantation.

**3.1 TAVI devices** 

Fig. 1. The Edwards SAPIEN Transcatheter aortic valve (Photo courtesy of Edwards Lifesciences)

Fig. 2. Transcatheter aortic valve crimped onto the introducing catheter (Photo courtesy of Edwards Lifesciences)

retrieval and greater accuracy when deployed. Examples include the DirectFlow (Direct Medical Flow Inc., Santa Rosa, CA, USA), Lotus (Boston Scientific Inc., Natick, MA, USA) and HLT (Heart Leaflet Technologies Inc., Maple Grove, MN, USA). Other systems incorporate features that allow fixation in the supracoronary aorta (Accurate, Symentis Inc., Lausanne, Switzerland; St Jude, St Jude Medical Inc., St. Paul, MN, USA) and anatomical guides to demonstrate the position of the native valve and coronary arteries, thus facilitating deployment (Engager, Medtronic Inc., USA; JenaClip, JenaValve Inc., Munich, Germany). At present, little is known about these novel valves in terms of their efficacy, safety, feasibility, and long-term durability. Further work is required to characterize their outcomes before they can be recommended for use in routine clinical practice.

## **3.2 Percutaneous access**

Initial reports of TAVI used a transseptal approach with access via the venous system (Cribier *et al.*, 2002). This route was technically challenging and not reproducible. Retrograde arterial approaches are now much more widely used. Access is gained typically through the femoral

Transcatheter Aortic Valve Implantation 231

Transapical TAVI involves inserting the valve device in an antegrade fashion through the anterolateral chest wall and apex of the left ventricle (Figure 4). This is done under general anesthesia with cardiopulmonary bypass (CPB) on standby. CPB is usually established through the femoral vessels if required. The site of the incision is determined by transthoracic echocardiography. After the chest cavity has been entered, the pericardium is opened and secured to the thoracic wall. Pacing wires are then attached to the myocardium to facilitate rapid ventricular pacing. Two purse-string sutures are inserted into the apex, and an introducer sheath is passed between them into the left ventricle. Guided by imaging, the implant is positioned across the native valve and then deployed using balloon inflation and rapid ventricular pacing. Postoperatively, the patient should be nursed in an intensive care unit for at least 24 hours. Transapical TAVI involves a thoracotomy, and for this reason is not recommended for patients with severe respiratory disease that precludes one-lung ventilation. In rare instances, a mini-sternotomy has been combined with retrograde transaortic TAVI, where the device is inserted via the ascending aorta (Latsios *et al.*, 2010).

Fig. 4. Trocar being antegradely introduced through the apex of the left ventricle (Photo

courtesy of Edwards Lifesciences)

This technique is reserved for patients with no other access sites.

**3.3 Transapical approach** 

artery (Webb *et al.*, 2006). The axillary, subclavian, and retroperitoneal iliac arteries, as well as the ascending aorta, have also been successfully used (Webb and Cribier, 2010). Cut-down to expose the arteries is sometimes done: this improves the ease of cannulation of the vessels and ensures safe closure. In such cases, the patient is usually anesthetized, which confers additional risks in this frail population. Percutaneous arterial puncture and suture closure is now the standard of care and can be completed safely under sedation (De Jaegere *et al.*, 2007; Vavuranakis *et al.*, 2010). After arterial puncture, aortography is done to characterize the coronary vessels, diseased valve, and aorta. Balloon valvuloplasty is then used to dilate the native valve under rapid ventricular pacing, which decreases cardiac output while the balloon is inflated. Between periods of rapid pacing, the blood pressure must be allowed to normalize. Intraoperative imaging, including aortography, transesophageal echocardiography, and fluoroscopy, is used to identify the optimal position for the new valve. Once this has been determined, the valve is deployed. Rapid ventricular pacing is required for balloon expandable devices (Figure 3), but not for self-expandable systems. Post-procedural echocardiography and aortography are done to check the position and function of the implant, the patency of the coronary vessels, and the presence of early complications (e.g. aortic regurgitation, paravalvular leak, aortic dissection, hemopericardium). It is recommended that patients are nursed postoperatively in the cardiac intensive care unit with invasive monitoring.

Fig. 3. Showing the retrograde approach and balloon inflation (black arrow) of the transcatheter aortic valve (white arrow) (Photo courtesy of Edwards Lifesciences)

#### **3.3 Transapical approach**

230 Aortic Valve

artery (Webb *et al.*, 2006). The axillary, subclavian, and retroperitoneal iliac arteries, as well as the ascending aorta, have also been successfully used (Webb and Cribier, 2010). Cut-down to expose the arteries is sometimes done: this improves the ease of cannulation of the vessels and ensures safe closure. In such cases, the patient is usually anesthetized, which confers additional risks in this frail population. Percutaneous arterial puncture and suture closure is now the standard of care and can be completed safely under sedation (De Jaegere *et al.*, 2007; Vavuranakis *et al.*, 2010). After arterial puncture, aortography is done to characterize the coronary vessels, diseased valve, and aorta. Balloon valvuloplasty is then used to dilate the native valve under rapid ventricular pacing, which decreases cardiac output while the balloon is inflated. Between periods of rapid pacing, the blood pressure must be allowed to normalize. Intraoperative imaging, including aortography, transesophageal echocardiography, and fluoroscopy, is used to identify the optimal position for the new valve. Once this has been determined, the valve is deployed. Rapid ventricular pacing is required for balloon expandable devices (Figure 3), but not for self-expandable systems. Post-procedural echocardiography and aortography are done to check the position and function of the implant, the patency of the coronary vessels, and the presence of early complications (e.g. aortic regurgitation, paravalvular leak, aortic dissection, hemopericardium). It is recommended that patients are nursed postoperatively in the cardiac intensive care unit with invasive monitoring.

Fig. 3. Showing the retrograde approach and balloon inflation (black arrow) of the trans-

catheter aortic valve (white arrow) (Photo courtesy of Edwards Lifesciences)

Transapical TAVI involves inserting the valve device in an antegrade fashion through the anterolateral chest wall and apex of the left ventricle (Figure 4). This is done under general anesthesia with cardiopulmonary bypass (CPB) on standby. CPB is usually established through the femoral vessels if required. The site of the incision is determined by transthoracic echocardiography. After the chest cavity has been entered, the pericardium is opened and secured to the thoracic wall. Pacing wires are then attached to the myocardium to facilitate rapid ventricular pacing. Two purse-string sutures are inserted into the apex, and an introducer sheath is passed between them into the left ventricle. Guided by imaging, the implant is positioned across the native valve and then deployed using balloon inflation and rapid ventricular pacing. Postoperatively, the patient should be nursed in an intensive care unit for at least 24 hours. Transapical TAVI involves a thoracotomy, and for this reason is not recommended for patients with severe respiratory disease that precludes one-lung ventilation. In rare instances, a mini-sternotomy has been combined with retrograde transaortic TAVI, where the device is inserted via the ascending aorta (Latsios *et al.*, 2010). This technique is reserved for patients with no other access sites.

Fig. 4. Trocar being antegradely introduced through the apex of the left ventricle (Photo courtesy of Edwards Lifesciences)

events.

p < 0.001).

Transcatheter Aortic Valve Implantation 233

introduction of a procedure can be unethical because of a loss of equipoise. It is difficult to blind participants and clinicians in surgical RCTs due to differences in outward appearances of wounds, and because the surgeon will always know which procedure has been performed. This problem can be overcome by blinding outcome assessors and data analysts to the allocation sequence. Another problem with surgical RCTs relates to the fact that they are often costly to undertake, and follow-up needs to be long-term to identify late and rare

The PARTNER Trial was the first RCT to compare outcomes between TAVI and other interventions for severe AS (Leon *et al.*, 2010). The study consisted of two parallel, prospective, multicenter, randomized trials. The first of these (Cohort A) randomized to either TAVI or open AVR participants with severe AS who were considered high-risk for surgery (STS-PROM > 10% mortality risk or > 15% predicted 30-day mortality). In Cohort B, patients with severe AS and considered unsuitable for surgery (> 50% predicted 30-day mortality or a serious irreversible condition) were allocated to either TAVI or medical management, which included balloon valvuloplasty. The primary outcome measure in Cohort A was survival at one year. In Cohort B, the primary endpoint was initially survival for the duration of the study, although this was supplemented partway through the trial with a composite co-primary outcome of survival and time to first rehospitalization. Secondary outcome measures included: functional improvement in NYHA classification; freedom from major adverse cardiovascular and cerebrovascular events (MACCE); evidence of prosthetic valve dysfunction (hemolysis, infection, thrombosis, severe paravalvular leak, or migration); a six-minute walk test; length of hospital stay; total hospital days from the index procedure to one year postoperatively; HRQL at 30 days, six months, and one year; improvement in aortic valve area; and a composite of survival, recurrent hospitalization,

and NYHA class. The eligibility criteria for the PARTNER Trial are listed in Table 2.

To date only results from Cohort B have been published, with the findings of Cohort A expected in late 2011. Between May 2007 and March 2009, 358 consecutive patients with severe AS who were considered unsuitable for surgery (Cohort B) were enrolled at 21 centers. Randomization allocated 179 patients to receive TAVI (Edwards SAPIEN device using the transfemoral approach) and 179 to be treated with medical management. Participants were followed-up for at least one year. The rate of death from any cause at one year post-randomization (primary endpoint) was 30.7% in those treated with TAVI and 50.7% in those treated with medical care alone (hazard ratio: 0.55; 95% confidence interval: 0.40 to 0.76; p < 0.001). The cardiovascular mortality rate one year after randomization was also lower in the TAVI group (20.5% vs. 44.6%; hazard ratio: 0.39; 95% confidence interval: 0.27 to 0.56; p < 0.001). Furthermore, the composite endpoint of rate of death from any cause and rehospitalization at one year (co-primary endpoint) was 42.7% with TAVI compared with 71.6% with medical care alone (hazard ratio: 0.46; 95% confidence interval: 0.35 to 0.59;

Complications were observed more frequently in the TAVI arm of the trial. There was a greater incidence of cerebrovascular events after TAVI at both 30 days after randomization (6.7% vs. 1.7%, p = 0.03) and at one year (10.6% vs. 4.5%, p = 0.04). Patients who received TAVI were also more likely to suffer major bleeding or vascular complications. Despite this, 30-day mortality was similar between groups. Patients treated with TAVI demonstrated marked improvements in functional outcomes: 74.8% of patients alive at one year in the TAVI arm were asymptomatic or had only mild symptoms (NYHA classes I or II), compared with 42.0% of surviving participants in the medical care alone group (p < 0.001). In addition,

#### **3.4 Service structure**

The performance of TAVI should be restricted to a small number of high-volume centers with readily available input from specialist cardiothoracic surgeons, interventional cardiologists, cardiac anesthetists, intensivists and perfusionists. The center must be proficient in dealing with both open and percutaneous valvular interventions in high-risk populations. Familiarity with the procedure and multidisciplinary management are likely to improve the rate of successful implantation and limit the number of complications. Furthermore, if complications do occur, they can be managed without the delay associated with transferring patients to another institution. Interventional cardiologists should have experience of a range of percutaneous valvular interventions, large-bore peripheral cannulation, and percutaneous suture closure. Cardiac surgeons should routinely perform complex open valvular procedures and be able to offer rescue or salvage surgery when complications arise. Bleeding from damaged peripheral vessels is not infrequent following TAVI, so it is also advantageous to also have onsite access to vascular surgeons and radiologists trained in either open or endovascular arterial repair.

## **4. Outcomes**

Outcomes are events that are either present or absent in study participants at specific time points after an intervention or exposure. They can be clinical, patient-reported, healthcare economic, composite, or surrogate. Studies investigating outcomes of interventions for aortic valve pathology concentrate on the safety, feasibility, efficacy, and durability of treatment options. Safety is typically assessed with clinical measures, such as operative morbidity and mortality. Feasibility describes whether the procedure can be accomplished successfully without recourse to alternative treatment. Efficacy is defined as whether an intervention works in those who receive it. For aortic valve therapies, it is usually based on echocardiographic findings and functional outcomes, such as the New York Heart Association (NYHA) classification and HRQL tools. Durability includes long-term outcomes such as prosthesis failure, reoperation, and survival. High quality studies, including randomized controlled trials (RCTs), are required to evaluate these outcomes in patients with severe AS and to inform treatment choices between open AVR, TAVI, balloon valvuloplasty, and medical management. This section will summarize outcomes from studies that have investigated the use of TAVI.

#### **4.1 Evidence from the PARTNER Trial**

RCTs are the gold standard study design for assessing surgical innovation. However, there is a paucity of well-designed surgical RCTs, which is in part due to specific methodological difficulties. Surgery is a complex intervention and is comprised of multiple events that interact together to affect outcomes. For example, perioperative mortality may be affected by patient factors (e.g. comorbidities), surgeon factors (e.g. skill and technique), anesthetic factors (e.g. quality of postoperative care) and service factors (e.g. number of nursing staff, rehabilitation services). If these factors are poorly controlled in an RCT, then confounding variables may result in bias. It is important that trials clearly predefine all aspects of the intervention in the study protocol and report protocol deviations in subsequent publications. The timing of surgical RCTs is also critical. RCTs undertaken too early in the development of a novel intervention may underestimate treatment effect magnitude as a consequence of operator learning-curve effects. RCTs undertaken too late after the

The performance of TAVI should be restricted to a small number of high-volume centers with readily available input from specialist cardiothoracic surgeons, interventional cardiologists, cardiac anesthetists, intensivists and perfusionists. The center must be proficient in dealing with both open and percutaneous valvular interventions in high-risk populations. Familiarity with the procedure and multidisciplinary management are likely to improve the rate of successful implantation and limit the number of complications. Furthermore, if complications do occur, they can be managed without the delay associated with transferring patients to another institution. Interventional cardiologists should have experience of a range of percutaneous valvular interventions, large-bore peripheral cannulation, and percutaneous suture closure. Cardiac surgeons should routinely perform complex open valvular procedures and be able to offer rescue or salvage surgery when complications arise. Bleeding from damaged peripheral vessels is not infrequent following TAVI, so it is also advantageous to also have onsite access to vascular surgeons and

Outcomes are events that are either present or absent in study participants at specific time points after an intervention or exposure. They can be clinical, patient-reported, healthcare economic, composite, or surrogate. Studies investigating outcomes of interventions for aortic valve pathology concentrate on the safety, feasibility, efficacy, and durability of treatment options. Safety is typically assessed with clinical measures, such as operative morbidity and mortality. Feasibility describes whether the procedure can be accomplished successfully without recourse to alternative treatment. Efficacy is defined as whether an intervention works in those who receive it. For aortic valve therapies, it is usually based on echocardiographic findings and functional outcomes, such as the New York Heart Association (NYHA) classification and HRQL tools. Durability includes long-term outcomes such as prosthesis failure, reoperation, and survival. High quality studies, including randomized controlled trials (RCTs), are required to evaluate these outcomes in patients with severe AS and to inform treatment choices between open AVR, TAVI, balloon valvuloplasty, and medical management. This section will summarize outcomes from

RCTs are the gold standard study design for assessing surgical innovation. However, there is a paucity of well-designed surgical RCTs, which is in part due to specific methodological difficulties. Surgery is a complex intervention and is comprised of multiple events that interact together to affect outcomes. For example, perioperative mortality may be affected by patient factors (e.g. comorbidities), surgeon factors (e.g. skill and technique), anesthetic factors (e.g. quality of postoperative care) and service factors (e.g. number of nursing staff, rehabilitation services). If these factors are poorly controlled in an RCT, then confounding variables may result in bias. It is important that trials clearly predefine all aspects of the intervention in the study protocol and report protocol deviations in subsequent publications. The timing of surgical RCTs is also critical. RCTs undertaken too early in the development of a novel intervention may underestimate treatment effect magnitude as a consequence of operator learning-curve effects. RCTs undertaken too late after the

radiologists trained in either open or endovascular arterial repair.

studies that have investigated the use of TAVI.

**4.1 Evidence from the PARTNER Trial** 

**3.4 Service structure** 

**4. Outcomes** 

introduction of a procedure can be unethical because of a loss of equipoise. It is difficult to blind participants and clinicians in surgical RCTs due to differences in outward appearances of wounds, and because the surgeon will always know which procedure has been performed. This problem can be overcome by blinding outcome assessors and data analysts to the allocation sequence. Another problem with surgical RCTs relates to the fact that they are often costly to undertake, and follow-up needs to be long-term to identify late and rare events.

The PARTNER Trial was the first RCT to compare outcomes between TAVI and other interventions for severe AS (Leon *et al.*, 2010). The study consisted of two parallel, prospective, multicenter, randomized trials. The first of these (Cohort A) randomized to either TAVI or open AVR participants with severe AS who were considered high-risk for surgery (STS-PROM > 10% mortality risk or > 15% predicted 30-day mortality). In Cohort B, patients with severe AS and considered unsuitable for surgery (> 50% predicted 30-day mortality or a serious irreversible condition) were allocated to either TAVI or medical management, which included balloon valvuloplasty. The primary outcome measure in Cohort A was survival at one year. In Cohort B, the primary endpoint was initially survival for the duration of the study, although this was supplemented partway through the trial with a composite co-primary outcome of survival and time to first rehospitalization. Secondary outcome measures included: functional improvement in NYHA classification; freedom from major adverse cardiovascular and cerebrovascular events (MACCE); evidence of prosthetic valve dysfunction (hemolysis, infection, thrombosis, severe paravalvular leak, or migration); a six-minute walk test; length of hospital stay; total hospital days from the index procedure to one year postoperatively; HRQL at 30 days, six months, and one year; improvement in aortic valve area; and a composite of survival, recurrent hospitalization, and NYHA class. The eligibility criteria for the PARTNER Trial are listed in Table 2.

To date only results from Cohort B have been published, with the findings of Cohort A expected in late 2011. Between May 2007 and March 2009, 358 consecutive patients with severe AS who were considered unsuitable for surgery (Cohort B) were enrolled at 21 centers. Randomization allocated 179 patients to receive TAVI (Edwards SAPIEN device using the transfemoral approach) and 179 to be treated with medical management. Participants were followed-up for at least one year. The rate of death from any cause at one year post-randomization (primary endpoint) was 30.7% in those treated with TAVI and 50.7% in those treated with medical care alone (hazard ratio: 0.55; 95% confidence interval: 0.40 to 0.76; p < 0.001). The cardiovascular mortality rate one year after randomization was also lower in the TAVI group (20.5% vs. 44.6%; hazard ratio: 0.39; 95% confidence interval: 0.27 to 0.56; p < 0.001). Furthermore, the composite endpoint of rate of death from any cause and rehospitalization at one year (co-primary endpoint) was 42.7% with TAVI compared with 71.6% with medical care alone (hazard ratio: 0.46; 95% confidence interval: 0.35 to 0.59; p < 0.001).

Complications were observed more frequently in the TAVI arm of the trial. There was a greater incidence of cerebrovascular events after TAVI at both 30 days after randomization (6.7% vs. 1.7%, p = 0.03) and at one year (10.6% vs. 4.5%, p = 0.04). Patients who received TAVI were also more likely to suffer major bleeding or vascular complications. Despite this, 30-day mortality was similar between groups. Patients treated with TAVI demonstrated marked improvements in functional outcomes: 74.8% of patients alive at one year in the TAVI arm were asymptomatic or had only mild symptoms (NYHA classes I or II), compared with 42.0% of surviving participants in the medical care alone group (p < 0.001). In addition,

Transcatheter Aortic Valve Implantation 235

• Native aortic annulus size < 18 mm or > 25 mm as measured by echocardiogram. • Recent (within 6 months) cerebrovascular accident or transient ischemic attack. • Renal insufficiency (creatinine > 3.0 mg/dL) or end-stage renal disease requiring

• Significant abdominal or thoracic aorta disease, including aneurysm (defined as maximal luminal diameter 5 cm or greater), marked tortuosity (hyperacute bend), aortic arch atheroma (especially if thick [> 5 mm], protruding, or ulcerated), narrowing of the abdominal aorta (especially with calcification and surface

• Iliofemoral vessel characteristics that would preclude safe placement of 22F or 24F introducer sheath, such as severe calcification, severe tortuosity or vessels size

Table 2. Eligibility criteria for participants included in the PARTNER Trial (Leon *et al.*, 2010) there was a significant improvement in the six-minute walk test in patients treated with TAVI, but not in those who received medical care alone. Echocardiography demonstrated a significant reduction in aortic valve area and transvalvular gradient at 30 days in patients receiving TAVI. Moreover, these findings were maintained at one year, which suggested

Leon *et al.* (2010) concluded that medical care alone did not alter the course of severe AS in patients who were not candidates for surgery. Transfemoral TAVI was markedly superior to medical care alone in this high-risk cohort of patients, and TAVI improved the rates of allcause and cardiovascular mortality. This is emphasized by the fact that only five patients needed to be treated with TAVI to prevent one death in the first year of follow-up. The safety of TAVI was highlighted by the similar 30-day mortality rate to medical care alone, despite a greater risk of vascular damage and bleeding, which were attributed to the large bore femoral access sheaths required in early versions of the Edwards SAPIEN heart-valve system. It is likely that future use of lower profile sheaths will reduce the incidence of vascular damage. Stroke rates were greater with TAVI and are the likely consequence of atherosclerotic emboli released during deployment of the valve. Less traumatic TAVI systems and novel cerebrovascular protection devices may help limit the incidence of stroke. It is noteworthy that increased survival was associated with improved function: patients treated with TAVI not only lived longer but also had fewer symptoms. Transfemoral TAVI is the current standard of care in patients who are not considered candidates for open AVR. The PARTNER Trial provides the best evidence yet to support the use of TAVI, although it is important to interpret the findings of Cohort B in the light of several limitations. The first of these relates to external validity: the PARTNER Trial was predominantly explanatory rather than pragmatic and had strict eligibility criteria (Table 2). Consequently, the results should not be extrapolated to patients with characteristics different from those enrolled in the trial, such as patients with peripheral vascular disease or severe LV dysfunction and those requiring coronary artery bypass grafts. Furthermore, the trial investigated only the transfemoral approach with a single device (Edwards SAPIEN), which has now been superseded by newer prostheses. Methodologically the trial is limited by poor reporting of randomization, allocation concealment, and blinding, while there is evidence of selective

irregularities), or severe "unfolding" and tortuosity of the thoracic aorta.

• Life expectancy < 12 months due to noncardiac comorbid conditions.

diameter < 7 mm for 22F sheath or < 8 mm for 24F sheath.

that TAVI devices are durable at least into the medium term.

premedicated.

chronic dialysis.

or clopidogrel (Plavix), or sensitivity to contrast media, which cannot be adequately

### **Inclusion criteria for the PARTNER Trial:**


Cohort B All candidates for Cohort B in this study must meet #2, 3, 4, 5 of the above criteria and:

• The patient, after formal consults by a cardiologist and two cardiovascular surgeons, agrees that medical factors preclude operation, based on a conclusion that the probability of death or serious, irreversible morbidity exceeds the probability of meaningful improvement. Specifically, the probability of death or serious, irreversible morbidity should exceed 50%.

#### **Exclusion Criteria for the PARTNER Trial:**


• Patients must have comorbidities such that the surgeon and cardiologist Co-Principle Investigators concur that the predicted risk of operative mortality is ≥ 15% and/ or a

• Patient has senile degenerative aortic valve stenosis with echocardiographically derived criteria: mean gradient > 40 mmHg or jet velocity greater than 4.0 m/s or an

• Patient is symptomatic from his/her aortic valve stenosis, as demonstrated by

• The patient or the patient's legal representative has been informed of the nature of the study, agrees to its provisions, and has provided written informed consent as

• The patient and the treating physician agree that the patient will return for all

Cohort B All candidates for Cohort B in this study must meet #2, 3, 4, 5 of the above

• The patient, after formal consults by a cardiologist and two cardiovascular surgeons, agrees that medical factors preclude operation, based on a conclusion that the probability of death or serious, irreversible morbidity exceeds the probability of meaningful improvement. Specifically, the probability of death or serious,

• Evidence of an acute myocardial infarction ≤ 1 month before the intended treatment (defined as Q wave MI, or non-Q wave MI with total CK elevation ≥ twice normal in the presence of CK-MB elevation or troponin level elevation (WHO definition). • Aortic valve was a congenital unicuspid or congenital bicuspid valve, or was

• Mixed aortic valve disease (aortic stenosis and aortic regurgitation with predominant

• Any therapeutic invasive cardiac procedure performed within 30 days of the index procedure, (or 6 months if the procedure was a drug eluting coronary stent

• Pre-existing prosthetic heart valve in any position, prosthetic ring, severe mitral

• Blood dyscrasias as defined: leukopenia (WBC < 3000 mm3); acute anemia (Hb < 9 mg%); thrombocytopenia (platelet count < 50,000 cells/mm3); history of bleeding

• Untreated clinically significant coronary artery disease requiring revascularization. • Hemodynamic instability requiring inotropic therapy or mechanical hemodynamic

annular calcification, or severe (greater than 3+) mitral regurgitation.

• Echocardiographic evidence of intracardiac mass, thrombus or vegetation. • Active peptic ulcer or upper gastrointestinal bleeding within the prior 3 months. • A known hypersensitivity or contraindication to aspirin, heparin, ticlopidine (Ticlid),

**Inclusion criteria for the PARTNER Trial:** 

minimum STS-PROM score of 10.

initial aortic valve area of < 0.8 cm2.

NYHA Functional Class II or greater.

required post-procedure follow-up visits.

irreversible morbidity should exceed 50%. **Exclusion Criteria for the PARTNER Trial:** 

criteria and:

noncalcified.

implantation).

support devices.

aortic regurgitation > 3+).

diathesis or coagulopathy.

• Need for emergency surgery for any reason.

• Severe ventricular dysfunction with LVEF < 20%.

• Hypertrophic cardiomyopathy with or without obstruction.

approved by the IRB of the respective clinical site.

or clopidogrel (Plavix), or sensitivity to contrast media, which cannot be adequately premedicated.


Table 2. Eligibility criteria for participants included in the PARTNER Trial (Leon *et al.*, 2010)

there was a significant improvement in the six-minute walk test in patients treated with TAVI, but not in those who received medical care alone. Echocardiography demonstrated a significant reduction in aortic valve area and transvalvular gradient at 30 days in patients receiving TAVI. Moreover, these findings were maintained at one year, which suggested that TAVI devices are durable at least into the medium term.

Leon *et al.* (2010) concluded that medical care alone did not alter the course of severe AS in patients who were not candidates for surgery. Transfemoral TAVI was markedly superior to medical care alone in this high-risk cohort of patients, and TAVI improved the rates of allcause and cardiovascular mortality. This is emphasized by the fact that only five patients needed to be treated with TAVI to prevent one death in the first year of follow-up. The safety of TAVI was highlighted by the similar 30-day mortality rate to medical care alone, despite a greater risk of vascular damage and bleeding, which were attributed to the large bore femoral access sheaths required in early versions of the Edwards SAPIEN heart-valve system. It is likely that future use of lower profile sheaths will reduce the incidence of vascular damage. Stroke rates were greater with TAVI and are the likely consequence of atherosclerotic emboli released during deployment of the valve. Less traumatic TAVI systems and novel cerebrovascular protection devices may help limit the incidence of stroke. It is noteworthy that increased survival was associated with improved function: patients treated with TAVI not only lived longer but also had fewer symptoms. Transfemoral TAVI is the current standard of care in patients who are not considered candidates for open AVR.

The PARTNER Trial provides the best evidence yet to support the use of TAVI, although it is important to interpret the findings of Cohort B in the light of several limitations. The first of these relates to external validity: the PARTNER Trial was predominantly explanatory rather than pragmatic and had strict eligibility criteria (Table 2). Consequently, the results should not be extrapolated to patients with characteristics different from those enrolled in the trial, such as patients with peripheral vascular disease or severe LV dysfunction and those requiring coronary artery bypass grafts. Furthermore, the trial investigated only the transfemoral approach with a single device (Edwards SAPIEN), which has now been superseded by newer prostheses. Methodologically the trial is limited by poor reporting of randomization, allocation concealment, and blinding, while there is evidence of selective

Transcatheter Aortic Valve Implantation 237

have failed, the etiological factors were valve embolization (n = 3), valve migration (n = 1),

The Medtronic CoreValve system is licensed for use using only retrograde percutaneous methods, typically transfemoral. The feasibility outcomes for this device are excellent, with several studies reporting 100% success rates (Behan *et al.*, 2008; Berry *et al.*, 2007; De Jaegere *et al.*, 2008). In a large prospective registry of 646 patients by Piazza *et al.* (2008), the procedural success rate was 97%. The authors did not present reasons for failure of implantation of the CoreValve device. In a small study, Grube *et al.* (2006) achieved successful implantation of TAVI prostheses in 21 out of 25 patients (84%). Reasons for procedural failure included paravalvular leak (n = 2), inability to cross a heavily calcified aortic valve (n = 1), and sudden death during balloon valvuloplasty (n = 1). In a second multicenter study by Grube et al. (2007), the procedural success rate was comparatively low (74%), which appears to be the consequence of malpositioning of a significant number of implants. Tamburino *et al.* (2009) reported outcomes for 30 patients treated with CoreValve TAVI. The procedural success rate was 93%, with one incident of pericardial tamponade and one of malpositioning of the TAVI device necessitating VIV implantation. It can be concluded from these reports that TAVI is a feasible procedure and that success rates are

likely to improve with greater operator experience and more advanced devices.

different devices and between transapical and transfemoral access routes.

covered stents and balloon tamponade to control bleeding vessels.

The assessment of safety in cardiothoracic surgery is made through reporting 30-day major adverse cardiovascular and cerebrovascular events (MACCE). Thirty-day mortality is conventionally defined as the occurrence of death from any cause within 30 days of a procedure. However, definitions are sometimes heterogeneous and must be considered carefully when outcomes from multiple studies are combined. In the case of TAVI, 30-day mortality is generally favorable and ranges from 0% to 25% (Behan *et al.*, 2008; Berry *et al.*, 2007; Cribier *et al.*, 2002; De Jaegere *et al.*, 2008; Descoutures *et al.*, 2008; Grube *et al.*, 2006; Grube *et al.*, 2007; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Svensson *et al.*, 2008; Tamburino *et al.*, 2009; Walther *et al.*, 2007; Walther *et al.*, 2008; Webb *et al.*, 2007; Ye *et al.*, 2009; Zierer *et al.*, 2008). Thirty-day mortality rates appear to be similar between the

Vascular complications are one of the major concerns with percutaneous approaches. Vascular injury has been shown to occur in up to 18% of TAVI procedures and can lead to hemorrhage, limb ischemia, and amputation (Behan *et al.*, 2008; Descoutures *et al.*, 2008; Leon *et al.*, 2010; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Tamburino *et al.*, 2009; Thomas *et al.*, 2010; Webb *et al.*, 2007; Zierer *et al.*, 2008). The etiology of vascular damage is often attributed to the large-caliber sheaths used with early TAVI devices. It is envisaged that the introduction of low-profile introducers and greater operator experience will reduce vascular complications. In addition, percutaneous vessel closure devices for transfemoral access are now widely available and will contribute further to the reduction in periprocedural major hemorrhage. Onsite access to vascular surgeons and interventional radiologists with experience of open and endovascular repair of damaged vessels is encouraged. The team should be familiar with the use of crossover femoral cannulation,

Stroke and transient ischemic attacks (TIAs) are common sequelae of TAVI deployment (range: 0% to 10%) and are believed to be the consequence of atheromatous emboli from the

and paravalvular leak (n = 1).

**4.3 Safety of TAVI** 

reporting of outcomes and the introduction of a co-primary outcome during the conduct of the trial. One final observation is that the trial offers only limited information on the longterm durability of TAVI prostheses. Additional long-term randomized controlled trials are warranted, the results of which, together with the results of Cohort A, will offer some insight into the relative efficacy of TAVI and open AVR in high-risk patients.

Since the first report of TAVI almost a decade ago, there has been an explosion in the number of related publications. Most of these are retrospective case series or comparative studies contrasting different approaches (transapical vs. transfemoral) or devices (Medtronic CoreValve vs. Edwards SAPIEN). Although the publications do not provide the same level of evidence as RCTs and are open to selection and recall bias, it is worth considering their findings because of their number. The next section summarizes the support or otherwise for TAVI from nonrandomized studies in terms of feasibility, safety, efficacy, and durability.

#### **4.2 Feasibility of TAVI**

The feasibility of TAVI can be assessed by considering the procedural success rate, defined as whether the implant was successfully deployed without immediate complications or the need to convert to open surgery. Early reports of procedural success rates are likely to have been affected by operator learning-curve effects. Cribier *et al.* (2002) reported that, using the antegrade transvenous approach, 22 out of 26 implants (84.6%) were successfully deployed, with four failures due to technical complications. Failures occurred as a consequence of valve migration immediately after the procedure (n = 2) and poor tolerance of the guide wire across the mitral valve (n = 2). This high rate of technical achievement could not be replicated by other researchers using transvenous access due to the complexity of passing the guide wire through the interventricular septum and mitral valve.

The introduction of retrograde methods for accessing the diseased aortic valve has led to greater procedure reproducibility. Webb *et al.* (2007) initially reported outcomes for transfemoral TAVI using the Edwards prosthesis in 50 patients. The procedural success rate in this cohort was 86% (43 out of 50 patients), with failure associated with inability to pass the catheter through the iliac artery (n = 1) or across the aortic valve (n = 3), device malpositioning (n = 2), or malfunction of the delivery system (n = 1). Other authors have demonstrated similar success rates for the transfemoral approach using the Edwards Lifesciences devices. Rodes-Cabau *et al.* (2008) successfully implanted TAVI prostheses in 91% of their patients, with failure occurring as a result of severely calcified iliac disease (n = 1) and intra-operative death secondary to myocardial ischaemia (n = 1). Similarly, Descoutures *et al.* (2008) reported success in 10 out of 12 patients (83%). In this series, procedural failure was the consequence of an inability to cannulate the iliac vessels (n = 1) and of fatal hemopericardium due to left ventricular perforation by the guide wire (n = 1).

Only Edwards Lifesciences devices are currently licensed for use using the transapical approach. Ye *et al.* (2009) and Zierer *et al.* (2008) both reported success rates of 100% in small case series using transapical access. In larger studies the success rates are similarly impressive. In a study by Walther *et al.* (2008) successful implantation was accomplished in 47 out of 50 patients (94%), with three patients requiring conversion to open AVR. In another multicentre study by Walther *et al*, 55 out of 59 implants were deployed successfully with four patients requiring urgent sternotomy and AVR due to device malposition (Walther *et al.*, 2007). In an article by Svensson *et al.* (2008), procedural success was 88% in 40 patients undergoing transapical TAVI. Of the five patients in whom TAVI was deemed to

reporting of outcomes and the introduction of a co-primary outcome during the conduct of the trial. One final observation is that the trial offers only limited information on the longterm durability of TAVI prostheses. Additional long-term randomized controlled trials are warranted, the results of which, together with the results of Cohort A, will offer some

Since the first report of TAVI almost a decade ago, there has been an explosion in the number of related publications. Most of these are retrospective case series or comparative studies contrasting different approaches (transapical vs. transfemoral) or devices (Medtronic CoreValve vs. Edwards SAPIEN). Although the publications do not provide the same level of evidence as RCTs and are open to selection and recall bias, it is worth considering their findings because of their number. The next section summarizes the support or otherwise for TAVI from nonrandomized studies in terms of feasibility, safety, efficacy, and durability.

The feasibility of TAVI can be assessed by considering the procedural success rate, defined as whether the implant was successfully deployed without immediate complications or the need to convert to open surgery. Early reports of procedural success rates are likely to have been affected by operator learning-curve effects. Cribier *et al.* (2002) reported that, using the antegrade transvenous approach, 22 out of 26 implants (84.6%) were successfully deployed, with four failures due to technical complications. Failures occurred as a consequence of valve migration immediately after the procedure (n = 2) and poor tolerance of the guide wire across the mitral valve (n = 2). This high rate of technical achievement could not be replicated by other researchers using transvenous access due to the complexity of passing

The introduction of retrograde methods for accessing the diseased aortic valve has led to greater procedure reproducibility. Webb *et al.* (2007) initially reported outcomes for transfemoral TAVI using the Edwards prosthesis in 50 patients. The procedural success rate in this cohort was 86% (43 out of 50 patients), with failure associated with inability to pass the catheter through the iliac artery (n = 1) or across the aortic valve (n = 3), device malpositioning (n = 2), or malfunction of the delivery system (n = 1). Other authors have demonstrated similar success rates for the transfemoral approach using the Edwards Lifesciences devices. Rodes-Cabau *et al.* (2008) successfully implanted TAVI prostheses in 91% of their patients, with failure occurring as a result of severely calcified iliac disease (n = 1) and intra-operative death secondary to myocardial ischaemia (n = 1). Similarly, Descoutures *et al.* (2008) reported success in 10 out of 12 patients (83%). In this series, procedural failure was the consequence of an inability to cannulate the iliac vessels (n = 1) and of fatal

Only Edwards Lifesciences devices are currently licensed for use using the transapical approach. Ye *et al.* (2009) and Zierer *et al.* (2008) both reported success rates of 100% in small case series using transapical access. In larger studies the success rates are similarly impressive. In a study by Walther *et al.* (2008) successful implantation was accomplished in 47 out of 50 patients (94%), with three patients requiring conversion to open AVR. In another multicentre study by Walther *et al*, 55 out of 59 implants were deployed successfully with four patients requiring urgent sternotomy and AVR due to device malposition (Walther *et al.*, 2007). In an article by Svensson *et al.* (2008), procedural success was 88% in 40 patients undergoing transapical TAVI. Of the five patients in whom TAVI was deemed to

insight into the relative efficacy of TAVI and open AVR in high-risk patients.

the guide wire through the interventricular septum and mitral valve.

hemopericardium due to left ventricular perforation by the guide wire (n = 1).

**4.2 Feasibility of TAVI** 

have failed, the etiological factors were valve embolization (n = 3), valve migration (n = 1), and paravalvular leak (n = 1).

The Medtronic CoreValve system is licensed for use using only retrograde percutaneous methods, typically transfemoral. The feasibility outcomes for this device are excellent, with several studies reporting 100% success rates (Behan *et al.*, 2008; Berry *et al.*, 2007; De Jaegere *et al.*, 2008). In a large prospective registry of 646 patients by Piazza *et al.* (2008), the procedural success rate was 97%. The authors did not present reasons for failure of implantation of the CoreValve device. In a small study, Grube *et al.* (2006) achieved successful implantation of TAVI prostheses in 21 out of 25 patients (84%). Reasons for procedural failure included paravalvular leak (n = 2), inability to cross a heavily calcified aortic valve (n = 1), and sudden death during balloon valvuloplasty (n = 1). In a second multicenter study by Grube et al. (2007), the procedural success rate was comparatively low (74%), which appears to be the consequence of malpositioning of a significant number of implants. Tamburino *et al.* (2009) reported outcomes for 30 patients treated with CoreValve TAVI. The procedural success rate was 93%, with one incident of pericardial tamponade and one of malpositioning of the TAVI device necessitating VIV implantation. It can be concluded from these reports that TAVI is a feasible procedure and that success rates are likely to improve with greater operator experience and more advanced devices.

#### **4.3 Safety of TAVI**

The assessment of safety in cardiothoracic surgery is made through reporting 30-day major adverse cardiovascular and cerebrovascular events (MACCE). Thirty-day mortality is conventionally defined as the occurrence of death from any cause within 30 days of a procedure. However, definitions are sometimes heterogeneous and must be considered carefully when outcomes from multiple studies are combined. In the case of TAVI, 30-day mortality is generally favorable and ranges from 0% to 25% (Behan *et al.*, 2008; Berry *et al.*, 2007; Cribier *et al.*, 2002; De Jaegere *et al.*, 2008; Descoutures *et al.*, 2008; Grube *et al.*, 2006; Grube *et al.*, 2007; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Svensson *et al.*, 2008; Tamburino *et al.*, 2009; Walther *et al.*, 2007; Walther *et al.*, 2008; Webb *et al.*, 2007; Ye *et al.*, 2009; Zierer *et al.*, 2008). Thirty-day mortality rates appear to be similar between the different devices and between transapical and transfemoral access routes.

Vascular complications are one of the major concerns with percutaneous approaches. Vascular injury has been shown to occur in up to 18% of TAVI procedures and can lead to hemorrhage, limb ischemia, and amputation (Behan *et al.*, 2008; Descoutures *et al.*, 2008; Leon *et al.*, 2010; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Tamburino *et al.*, 2009; Thomas *et al.*, 2010; Webb *et al.*, 2007; Zierer *et al.*, 2008). The etiology of vascular damage is often attributed to the large-caliber sheaths used with early TAVI devices. It is envisaged that the introduction of low-profile introducers and greater operator experience will reduce vascular complications. In addition, percutaneous vessel closure devices for transfemoral access are now widely available and will contribute further to the reduction in periprocedural major hemorrhage. Onsite access to vascular surgeons and interventional radiologists with experience of open and endovascular repair of damaged vessels is encouraged. The team should be familiar with the use of crossover femoral cannulation, covered stents and balloon tamponade to control bleeding vessels.

Stroke and transient ischemic attacks (TIAs) are common sequelae of TAVI deployment (range: 0% to 10%) and are believed to be the consequence of atheromatous emboli from the

Transcatheter Aortic Valve Implantation 239

transfusion, hypertension, chronic obstructive pulmonary disease, and transapical access (Bagur *et al.*, 2010). In patients with chronic kidney disease, the incidence of acute renal impairment is lower in those treated with TAVI than with open AVR (9% vs. 26%, p < 0.001) (Bagur *et al.*, 2010), which perhaps reflects the deleterious effects of cardiopulmonary bypass, hypotension, and ischemia associated with open surgery. TAVI may therefore be a

Open AVR in the presence of severe LV dysfunction is high-risk, and TAVI may be an appropriate alternative in this situation. A recent nonrandomized study (Clavel *et al.*, 2010) compared TAVI (n = 83) to open AVR (n = 200). Despite a higher STS-PROM score in patients who received TAVI, the authors reported that TAVI was associated with a greater improvement in ejection fraction than open AVR (14% vs. 7%; p < 0.001) and better hemodynamics at one year. Evidence from randomized trials is required to assess whether

Other complications associated with TAVI include: supraventricular tachyarrhythmia (range: 5% to 31%); ventricular tachyarrhythmia (range: 0% to 4%); myocardial infarction (range: 0% to 15%); cardiac tamponade (range: 2% to 10%); conversion to open surgery (range: 0% to 8%); conversion to valvuloplasty (range: 0% to 4%); emergency valve-in-valve procedure (range: 2% to 12%); and aortic dissection or rupture (range: 0% to 4%) (Behan *et al.*, 2008; Berry *et al.*, 2007; Cribier *et al.*, 2002; De Jaegere *et al.*, 2008; Descoutures *et al.*, 2008; Grube *et al.*, 2006; Grube *et al.*, 2007; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Svensson *et al.*, 2008; Tamburino *et al.*, 2009; Walther *et al.*, 2007; Walther *et al.*, 2008; Webb *et al.*, 2007; Ye *et al.*, 2009; Zierer *et al.*, 2008). The overall 30-day MACCE ranges from 3% to 35% (Behan *et al.*, 2008; Berry *et al.*, 2007; Cribier *et al.*, 2002; De Jaegere *et al.*, 2008; Descoutures *et al.*, 2008; Grube *et al.*, 2006; Grube *et al.*, 2007; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Svensson *et al.*, 2008; Tamburino *et al.*, 2009; Walther *et al.*,

The efficacy of valvular procedures can be determined by whether they improve echocardiographic measurements of hemodynamic performance and by the effect of treatment on patient function and quality of life. The main pathological findings at echocardiography in patients with severe AS are reduced AVA, raised peak and mean pressure gradients across the aortic valve, reduced LV ejection fraction, and LV dysfunction. Numerous studies have investigated the echocardiographic outcomes of TAVI and have consistently demonstrated statistically significant (p < 0.05) improvements in AVA, mean and peak aortic valve pressure gradients, and LV ejection fraction between preoperative and early postoperative values (Behan *et al.*, 2008; Berry *et al.*, 2007; Clavel *et al.*, 2009; Cribier *et al.*, 2002; De Jaegere *et al.*, 2008; Descoutures *et al.*, 2008; Figulla *et al.*, 2011; Grube *et al.*, 2006; Grube *et al.*, 2007; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Svensson *et al.*, 2008; Tamburino *et al.*, 2009; Walther *et al.*, 2007; Walther *et al.*, 2008; Webb *et al.*, 2007; Ye *et al.*, 2009; Zierer *et al.*, 2008). Furthermore, there is no deterioration in echocardiographic outcomes in patients followed-up for at least a year, which suggests that TAVI produces longer-lasting effects than balloon valvuloplasty alone (Cribier *et al.*, 2002; Figulla *et al.*, 2011;

It is imperative that improvements in hemodynamics translate into tangible benefits to patient function and health status. In studies of cardiothoracic surgery, patient function and

safer therapeutic option for AS patients with a history of chronic renal impairment.

TAVI results in better recovery of LV function than does AVR.

2007; Walther *et al.*, 2008; Webb *et al.*, 2007; Ye *et al.*, 2009; Zierer *et al.*, 2008).

**4.4 Efficacy of TAVI** 

Webb *et al.*, 2007; Ye *et al.*, 2009).

ascending aorta and diseased aortic valve (Berry *et al.*, 2007; Cribier *et al.*, 2002; Descoutures *et al.*, 2008; Grube *et al.*, 2006; Grube *et al.*, 2007; Leon *et al.*, 2010; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Svensson *et al.*, 2008; Tamburino *et al.*, 2009; Thomas *et al.*, 2010; Walther *et al.*, 2007; Webb *et al.*, 2007; Ye *et al.*, 2009; Zierer *et al.*, 2008). The risk of cerebrovascular events is increased in TAVI patients with atrial fibrillation and in those in whom valve thrombosis has occurred. Diffusion-weighted magnetic resonance imaging has demonstrated new cerebral lesions in up to 91% of patients undergoing TAVI (Ghanem *et al.*, 2010; Kahlert *et al.*, 2010). Fortunately, these radiological images do not correlate with clinically observed neurological deficits, which suggests that ischemic brain injury is predominantly subclinical (Lefevre *et al.*, 2011; Webb *et al.*, 2009a). The introduction of less traumatic delivery devices may help to reduce the incidence of cerebrovascular events, while novel catheters that are designed to capture or deflect emboli are under evaluation (Nietlispach *et al.*, 2010).

Occlusion of the left main coronary ostium is a potentially fatal complication of TAVI insertion. The usual mechanism involves upward displacement of the native aortic valve leaflet such that it completely covers the coronary ostia. Rarely, the device itself can abut against the coronary ostia, which reduces blood flow to the myocardium. Low coronary origin (less than 12 mm superior to the aortic annulus on computed tomography) or shallow coronary sinuses are thought to predispose to left main coronary artery occlusion (Tops *et al.*, 2008; Webb, 2009b). The Medtronic CoreValve has a tampered proximal end, which is designed to prevent coronary occlusion.

Bradycardia requiring a permanent pacemaker is a frequent problem following TAVI. It results from pressure effects on the conduction pathways that pass through the membranous interventricular septum beneath the aortic valve. This is particularly common in patients with a pre-existing bundle branch or atrioventricular block. Several additional factors are believed to predispose to pacemaker insertion: advanced age, oversizing of the implant, and the depth of the implant within the LV outflow tract (Willson & Webb, 2011). The Medtronic CoreValve is considerably longer than the Edwards SAPIEN and is in contact with a larger area of the interventricular septum. The CoreValve device is associated with considerably higher rates of pacemaker insertion (range: 20% to 38%) (Elchaninoff *et al.*, 2011; Jilaihawi *et al.*, 2010; Piazza *et al.*, 2010; Zahn *et al.*, 2011) compared to the Edwards SAPIEN (range 3% to 10%) (Elchaninoff *et al.*, 2011; Thomas *et al.*, 2010; Webb *et al.*, 2009a).

Although valvular aortic regurgitation is rare after TAVI, paravalvular leak occurs more commonly and is moderate or severe in up to 15% of patients (Leon *et al.*, 2010; Sherif *et al.*, 2010; Webb *et al.*, 2009a; Zahn *et al.*, 2011). Leak occurs when there is an inadequate seal between the outer surface of the device and the aortic annulus, which allows blood to flow around the periphery of the prosthesis. This may occur if the implant is deployed either too proximally or too distally in relation to the plane of the aortic annulus; when the chosen device is undersized in relation to the aortic annulus; or if the prosthesis fails to expand completely. Acute paravalvular leak can be treated with balloon valvuloplasty, retrieval of the device (if possible), or VIV techniques. TAVI has been shown to be associated with a higher incidence of paravalvular leak than open AVR (12% vs. 1%) (Leon *et al.*, 2010).

Acute renal impairment, defined as a glomerular filtration rate reduction of greater than 25%, is associated with a four-fold increase in 30-day mortality following TAVI (Willson & Webb, 2011). Acute renal impairment and renal replacement therapy occur in 11% and 1.4% of TAVI patients, respectively, with risk factors including chronic kidney disease, blood

ascending aorta and diseased aortic valve (Berry *et al.*, 2007; Cribier *et al.*, 2002; Descoutures *et al.*, 2008; Grube *et al.*, 2006; Grube *et al.*, 2007; Leon *et al.*, 2010; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Svensson *et al.*, 2008; Tamburino *et al.*, 2009; Thomas *et al.*, 2010; Walther *et al.*, 2007; Webb *et al.*, 2007; Ye *et al.*, 2009; Zierer *et al.*, 2008). The risk of cerebrovascular events is increased in TAVI patients with atrial fibrillation and in those in whom valve thrombosis has occurred. Diffusion-weighted magnetic resonance imaging has demonstrated new cerebral lesions in up to 91% of patients undergoing TAVI (Ghanem *et al.*, 2010; Kahlert *et al.*, 2010). Fortunately, these radiological images do not correlate with clinically observed neurological deficits, which suggests that ischemic brain injury is predominantly subclinical (Lefevre *et al.*, 2011; Webb *et al.*, 2009a). The introduction of less traumatic delivery devices may help to reduce the incidence of cerebrovascular events, while novel catheters that are designed to capture or deflect emboli are under evaluation

Occlusion of the left main coronary ostium is a potentially fatal complication of TAVI insertion. The usual mechanism involves upward displacement of the native aortic valve leaflet such that it completely covers the coronary ostia. Rarely, the device itself can abut against the coronary ostia, which reduces blood flow to the myocardium. Low coronary origin (less than 12 mm superior to the aortic annulus on computed tomography) or shallow coronary sinuses are thought to predispose to left main coronary artery occlusion (Tops *et al.*, 2008; Webb, 2009b). The Medtronic CoreValve has a tampered proximal end, which is

Bradycardia requiring a permanent pacemaker is a frequent problem following TAVI. It results from pressure effects on the conduction pathways that pass through the membranous interventricular septum beneath the aortic valve. This is particularly common in patients with a pre-existing bundle branch or atrioventricular block. Several additional factors are believed to predispose to pacemaker insertion: advanced age, oversizing of the implant, and the depth of the implant within the LV outflow tract (Willson & Webb, 2011). The Medtronic CoreValve is considerably longer than the Edwards SAPIEN and is in contact with a larger area of the interventricular septum. The CoreValve device is associated with considerably higher rates of pacemaker insertion (range: 20% to 38%) (Elchaninoff *et al.*, 2011; Jilaihawi *et al.*, 2010; Piazza *et al.*, 2010; Zahn *et al.*, 2011) compared to the Edwards SAPIEN (range 3% to 10%) (Elchaninoff *et al.*, 2011; Thomas *et al.*, 2010; Webb *et al.*, 2009a). Although valvular aortic regurgitation is rare after TAVI, paravalvular leak occurs more commonly and is moderate or severe in up to 15% of patients (Leon *et al.*, 2010; Sherif *et al.*, 2010; Webb *et al.*, 2009a; Zahn *et al.*, 2011). Leak occurs when there is an inadequate seal between the outer surface of the device and the aortic annulus, which allows blood to flow around the periphery of the prosthesis. This may occur if the implant is deployed either too proximally or too distally in relation to the plane of the aortic annulus; when the chosen device is undersized in relation to the aortic annulus; or if the prosthesis fails to expand completely. Acute paravalvular leak can be treated with balloon valvuloplasty, retrieval of the device (if possible), or VIV techniques. TAVI has been shown to be associated with a

higher incidence of paravalvular leak than open AVR (12% vs. 1%) (Leon *et al.*, 2010).

Acute renal impairment, defined as a glomerular filtration rate reduction of greater than 25%, is associated with a four-fold increase in 30-day mortality following TAVI (Willson & Webb, 2011). Acute renal impairment and renal replacement therapy occur in 11% and 1.4% of TAVI patients, respectively, with risk factors including chronic kidney disease, blood

(Nietlispach *et al.*, 2010).

designed to prevent coronary occlusion.

transfusion, hypertension, chronic obstructive pulmonary disease, and transapical access (Bagur *et al.*, 2010). In patients with chronic kidney disease, the incidence of acute renal impairment is lower in those treated with TAVI than with open AVR (9% vs. 26%, p < 0.001) (Bagur *et al.*, 2010), which perhaps reflects the deleterious effects of cardiopulmonary bypass, hypotension, and ischemia associated with open surgery. TAVI may therefore be a safer therapeutic option for AS patients with a history of chronic renal impairment.

Open AVR in the presence of severe LV dysfunction is high-risk, and TAVI may be an appropriate alternative in this situation. A recent nonrandomized study (Clavel *et al.*, 2010) compared TAVI (n = 83) to open AVR (n = 200). Despite a higher STS-PROM score in patients who received TAVI, the authors reported that TAVI was associated with a greater improvement in ejection fraction than open AVR (14% vs. 7%; p < 0.001) and better hemodynamics at one year. Evidence from randomized trials is required to assess whether TAVI results in better recovery of LV function than does AVR.

Other complications associated with TAVI include: supraventricular tachyarrhythmia (range: 5% to 31%); ventricular tachyarrhythmia (range: 0% to 4%); myocardial infarction (range: 0% to 15%); cardiac tamponade (range: 2% to 10%); conversion to open surgery (range: 0% to 8%); conversion to valvuloplasty (range: 0% to 4%); emergency valve-in-valve procedure (range: 2% to 12%); and aortic dissection or rupture (range: 0% to 4%) (Behan *et al.*, 2008; Berry *et al.*, 2007; Cribier *et al.*, 2002; De Jaegere *et al.*, 2008; Descoutures *et al.*, 2008; Grube *et al.*, 2006; Grube *et al.*, 2007; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Svensson *et al.*, 2008; Tamburino *et al.*, 2009; Walther *et al.*, 2007; Walther *et al.*, 2008; Webb *et al.*, 2007; Ye *et al.*, 2009; Zierer *et al.*, 2008). The overall 30-day MACCE ranges from 3% to 35% (Behan *et al.*, 2008; Berry *et al.*, 2007; Cribier *et al.*, 2002; De Jaegere *et al.*, 2008; Descoutures *et al.*, 2008; Grube *et al.*, 2006; Grube *et al.*, 2007; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Svensson *et al.*, 2008; Tamburino *et al.*, 2009; Walther *et al.*, 2007; Walther *et al.*, 2008; Webb *et al.*, 2007; Ye *et al.*, 2009; Zierer *et al.*, 2008).

#### **4.4 Efficacy of TAVI**

The efficacy of valvular procedures can be determined by whether they improve echocardiographic measurements of hemodynamic performance and by the effect of treatment on patient function and quality of life. The main pathological findings at echocardiography in patients with severe AS are reduced AVA, raised peak and mean pressure gradients across the aortic valve, reduced LV ejection fraction, and LV dysfunction. Numerous studies have investigated the echocardiographic outcomes of TAVI and have consistently demonstrated statistically significant (p < 0.05) improvements in AVA, mean and peak aortic valve pressure gradients, and LV ejection fraction between preoperative and early postoperative values (Behan *et al.*, 2008; Berry *et al.*, 2007; Clavel *et al.*, 2009; Cribier *et al.*, 2002; De Jaegere *et al.*, 2008; Descoutures *et al.*, 2008; Figulla *et al.*, 2011; Grube *et al.*, 2006; Grube *et al.*, 2007; Piazza *et al.*, 2008; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Svensson *et al.*, 2008; Tamburino *et al.*, 2009; Walther *et al.*, 2007; Walther *et al.*, 2008; Webb *et al.*, 2007; Ye *et al.*, 2009; Zierer *et al.*, 2008). Furthermore, there is no deterioration in echocardiographic outcomes in patients followed-up for at least a year, which suggests that TAVI produces longer-lasting effects than balloon valvuloplasty alone (Cribier *et al.*, 2002; Figulla *et al.*, 2011; Webb *et al.*, 2007; Ye *et al.*, 2009).

It is imperative that improvements in hemodynamics translate into tangible benefits to patient function and health status. In studies of cardiothoracic surgery, patient function and

Transcatheter Aortic Valve Implantation 241

from multiple studies reported that mean one-year survival after TAVI was 75.9% (range: 64.1% to 87.0%) (Al-Attar *et al.*, 2009; Grube *et al.*, 2008; Himbert *et al.*, 2009; Kapadia *et al.*, 2009; Otten *et al.*, 2009; Rajani *et al.*, 2010; Rodes-Cabau *et al.*, 2010; Thielmann *et al.*, 2009; Walther *et al.*, 2010; Webb *et al.*, 2009a; Ye *et al.*, 2009). In contrast the mean one-year survival rate for patients treated with medical care alone was 62.4% (range: 40.0% to 84.8%; p < 0.01 vs. TAVI), revealing a 13.5% survival advantage in favor of TAVI at one year. This is in agreement with the PARTNER Trial, which reported a 20% survival advantage for TAVI (Leon *et al.*, 2010). Gurvitch *et al.* (2010) are one of the few groups to report outcomes for TAVI beyond one year of follow-up: In a cohort of 70 patients undergoing TAVI, they reported one-, two-, and three-year survival rates of 81%, 74%, and 61%, respectively, although patients who died within 30-days or in whom TAVI failed were excluded from the analysis. During the follow-up period, there were 30 late deaths, of which three were valverelated: two patients died from intracerebral hemorrhage secondary to supratherapeutic anticoagulation, and sudden death occurred in another patient who was found postmortem to have an overgrowth of fibrous connective tissue around the prosthesis. No deaths were directly related to valvular dysfunction, which is in agreement with other studies reporting

outcomes up to one year (Grube *et al.*, 2008; Rodes-Cabau., 2010; Webb *et al.*, 2009a).

undertaking TAVI.

**5. Future directions** 

In the review by Figulla *et al.* (2011), one-year survival following transfemoral TAVI (79.2%, range: 68.1% to 87.0%) was superior to transapical access (73.6%, range 60.0% to 78.0%, p = 0.04). Reduced survival in patients receiving transapical TAVI may be explained by the need for general anesthesia, thoracotomy, and cannulation of the left ventricular apex. It is important to note that most studies included in Figulla *et al.* were retrospective and nonrandomized, and consequently at risk of bias. RCTs are required to determine which method of gaining access to the diseased aortic valve is the most efficacious when

Very little is known about the risk of valvular degeneration with TAVI devices. *In vitro* testing of the latest generation of Edwards SAPIEN/XT and Medtronic CoreValve devices suggests that durability in excess of 10 years can be expected (Willson & Webb., 2010). Because of the proven efficacy of open AVR, without long-term *in vivo* data, it is very unlikely that TAVI devices will be licensed for use in younger patients without comorbidities. Indeed, freedom from reoperation for valvular degeneration is greater than 95% with modern surgical bioprostheses (Jamieson *et al.*, 1995). The results of long-term follow-up will be required to answer this clinical question, but it is unlikely that TAVI will

The evidence base for TAVI is rapidly evolving, and there has been a significant rise in the number of new publications over the last five years. Most eagerly awaited are the findings of Cohort A of the PARTNER Trial. This will provide insight into the comparative efficacy of TAVI and open AVR in high-risk patients with severe symptomatic AS. The results of this trial will have important implications for healthcare policy implementation and may mean greater financial provision for TAVI in high-risk patients. In addition, the two main techniques for accessing the aortic valve (transfemoral and transapical) will need to be compared in an RCT, especially given that pooled data suggests one-year survival is worse after transapical TAVI (Figulla *et al.*, 2011). If this is proved to be the case, then the transapical approach may

replace open AVR for the management of uncomplicated, severe AS.

be restricted to patients in whom the transfemoral route is contraindicated.

the severity of symptoms are most commonly assessed through changes in NYHA classification. This clinician-reported outcome assigns patients to one of four categories, ranging from no symptoms or limitations on ordinary physical activity (Class I) to severe symptoms at rest necessitating continuous bed rest (Class IV). TAVI has been shown consistently to improve NYHA classification, with between 50% and 100% of patients demonstrating an improvement of at least one grade in NYHA classification at one-month post-procedure (Cribier *et al.*, 2002; Gotzmann *et al.*, 2010; Grube *et al.*, 2006; Grube *et al.*, 2007; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Svensson *et al.*, 2008; Webb *et al.*, 2007; Ye *et al.*, 2009). The short duration of follow-up of most studies means that it is difficult to determine whether these benefits are sustained, but some publications have reported that functional improvements last at least a year (Leon *et al.*, 2010; Webb *et al.*, 2009a). NYHA classification is also correlated with performance in the 6-minute walk test (Demers *et al.*, 2001; Gotzmann *et al.*, 2010).

Patient-reported outcomes (PROs), including health-related quality of life, provide unbiased assessments of health status from the patient's perspective. Instruments (most often questionnaires) designed to capture these issues are called HRQL tools and are multidimensional, encompassing perceptions of physical, emotional, and social function, as well as assessing specific symptoms caused by the disease and treatment (Fayers & Hays., 2005). The development of symptoms of congestive cardiac failure confers a poor prognosis on patients with severe AS and is likely to significantly affect HRQL. It is important to determine whether TAVI has a beneficial effect on HRQL and symptom palliation, in addition to increasing survival. Several studies have reported HRQL outcomes in patients treated with TAVI. Gotzmann *et al*., (2010) reported that HRQL (assessed using the Minnesota Living with Heart Failure Questionnaire [MLHFQ]) was significantly better 30 days after TAVI when compared to baseline. This correlated with observed improvements in the 6-minute walk test and a reduction in serum beta-natriuretic peptide. Krane *et al.*  (2010) measured HRQL with the Medical Outcomes Study Short Form-36 (SF-36) in a cohort of 99 patients treated with TAVI. Physical functioning, bodily pain, general health, and vitality all improved significantly from baseline at the three-month follow-up. Scores for social functioning and mental health remained static, while only role-emotional functioning deteriorated after TAVI. Ussia *et al.* (2009) used the SF-12 to compare HRQL in patients before and after TAVI with aged matched population reference values. Preprocedural AS patients had markedly worse HRQL than the general population for both physical and mental function. Five months after TAVI, HRQL had improved dramatically and was similar to that in the general population. This suggests that recovery following TAVI does occur, but takes several months. In the REVIVAL II feasibility study, HRQL was measured using the SF-12 and the Kansas City Cardiomyopathy Questionnaire (KCCQ) at 30 days and 6 months after TAVI insertion in 75 patients (Reynolds *et al.*, 2008). At 30 days, a significant improvement in KCCQ score was observed, but not in the physical or mental functioning scales of the SF-12. However, by the 6-month follow-up, improvements in all facets of HRQL were evident. Taken together, these results provide strong evidence that TAVI improves HRQL in patients with severe AS, although it takes up to 6 months for HRQL to match that in the general population.

#### **4.5 Durability of TAVI**

TAVI is a relatively new intervention, and long-term outcome data is scarce beyond one year of follow-up. A recent systematic review (Figulla *et al.*, 2011) incorporating pooled data

the severity of symptoms are most commonly assessed through changes in NYHA classification. This clinician-reported outcome assigns patients to one of four categories, ranging from no symptoms or limitations on ordinary physical activity (Class I) to severe symptoms at rest necessitating continuous bed rest (Class IV). TAVI has been shown consistently to improve NYHA classification, with between 50% and 100% of patients demonstrating an improvement of at least one grade in NYHA classification at one-month post-procedure (Cribier *et al.*, 2002; Gotzmann *et al.*, 2010; Grube *et al.*, 2006; Grube *et al.*, 2007; Rodes-Cabau *et al.*, 2008; Spargias *et al.*, 2008; Svensson *et al.*, 2008; Webb *et al.*, 2007; Ye *et al.*, 2009). The short duration of follow-up of most studies means that it is difficult to determine whether these benefits are sustained, but some publications have reported that functional improvements last at least a year (Leon *et al.*, 2010; Webb *et al.*, 2009a). NYHA classification is also correlated with performance in the 6-minute walk test (Demers *et al.*,

Patient-reported outcomes (PROs), including health-related quality of life, provide unbiased assessments of health status from the patient's perspective. Instruments (most often questionnaires) designed to capture these issues are called HRQL tools and are multidimensional, encompassing perceptions of physical, emotional, and social function, as well as assessing specific symptoms caused by the disease and treatment (Fayers & Hays., 2005). The development of symptoms of congestive cardiac failure confers a poor prognosis on patients with severe AS and is likely to significantly affect HRQL. It is important to determine whether TAVI has a beneficial effect on HRQL and symptom palliation, in addition to increasing survival. Several studies have reported HRQL outcomes in patients treated with TAVI. Gotzmann *et al*., (2010) reported that HRQL (assessed using the Minnesota Living with Heart Failure Questionnaire [MLHFQ]) was significantly better 30 days after TAVI when compared to baseline. This correlated with observed improvements in the 6-minute walk test and a reduction in serum beta-natriuretic peptide. Krane *et al.*  (2010) measured HRQL with the Medical Outcomes Study Short Form-36 (SF-36) in a cohort of 99 patients treated with TAVI. Physical functioning, bodily pain, general health, and vitality all improved significantly from baseline at the three-month follow-up. Scores for social functioning and mental health remained static, while only role-emotional functioning deteriorated after TAVI. Ussia *et al.* (2009) used the SF-12 to compare HRQL in patients before and after TAVI with aged matched population reference values. Preprocedural AS patients had markedly worse HRQL than the general population for both physical and mental function. Five months after TAVI, HRQL had improved dramatically and was similar to that in the general population. This suggests that recovery following TAVI does occur, but takes several months. In the REVIVAL II feasibility study, HRQL was measured using the SF-12 and the Kansas City Cardiomyopathy Questionnaire (KCCQ) at 30 days and 6 months after TAVI insertion in 75 patients (Reynolds *et al.*, 2008). At 30 days, a significant improvement in KCCQ score was observed, but not in the physical or mental functioning scales of the SF-12. However, by the 6-month follow-up, improvements in all facets of HRQL were evident. Taken together, these results provide strong evidence that TAVI improves HRQL in patients with severe AS, although it takes up to 6 months for HRQL to match that

TAVI is a relatively new intervention, and long-term outcome data is scarce beyond one year of follow-up. A recent systematic review (Figulla *et al.*, 2011) incorporating pooled data

2001; Gotzmann *et al.*, 2010).

in the general population.

**4.5 Durability of TAVI** 

from multiple studies reported that mean one-year survival after TAVI was 75.9% (range: 64.1% to 87.0%) (Al-Attar *et al.*, 2009; Grube *et al.*, 2008; Himbert *et al.*, 2009; Kapadia *et al.*, 2009; Otten *et al.*, 2009; Rajani *et al.*, 2010; Rodes-Cabau *et al.*, 2010; Thielmann *et al.*, 2009; Walther *et al.*, 2010; Webb *et al.*, 2009a; Ye *et al.*, 2009). In contrast the mean one-year survival rate for patients treated with medical care alone was 62.4% (range: 40.0% to 84.8%; p < 0.01 vs. TAVI), revealing a 13.5% survival advantage in favor of TAVI at one year. This is in agreement with the PARTNER Trial, which reported a 20% survival advantage for TAVI (Leon *et al.*, 2010). Gurvitch *et al.* (2010) are one of the few groups to report outcomes for TAVI beyond one year of follow-up: In a cohort of 70 patients undergoing TAVI, they reported one-, two-, and three-year survival rates of 81%, 74%, and 61%, respectively, although patients who died within 30-days or in whom TAVI failed were excluded from the analysis. During the follow-up period, there were 30 late deaths, of which three were valverelated: two patients died from intracerebral hemorrhage secondary to supratherapeutic anticoagulation, and sudden death occurred in another patient who was found postmortem to have an overgrowth of fibrous connective tissue around the prosthesis. No deaths were directly related to valvular dysfunction, which is in agreement with other studies reporting outcomes up to one year (Grube *et al.*, 2008; Rodes-Cabau., 2010; Webb *et al.*, 2009a).

In the review by Figulla *et al.* (2011), one-year survival following transfemoral TAVI (79.2%, range: 68.1% to 87.0%) was superior to transapical access (73.6%, range 60.0% to 78.0%, p = 0.04). Reduced survival in patients receiving transapical TAVI may be explained by the need for general anesthesia, thoracotomy, and cannulation of the left ventricular apex. It is important to note that most studies included in Figulla *et al.* were retrospective and nonrandomized, and consequently at risk of bias. RCTs are required to determine which method of gaining access to the diseased aortic valve is the most efficacious when undertaking TAVI.

Very little is known about the risk of valvular degeneration with TAVI devices. *In vitro* testing of the latest generation of Edwards SAPIEN/XT and Medtronic CoreValve devices suggests that durability in excess of 10 years can be expected (Willson & Webb., 2010). Because of the proven efficacy of open AVR, without long-term *in vivo* data, it is very unlikely that TAVI devices will be licensed for use in younger patients without comorbidities. Indeed, freedom from reoperation for valvular degeneration is greater than 95% with modern surgical bioprostheses (Jamieson *et al.*, 1995). The results of long-term follow-up will be required to answer this clinical question, but it is unlikely that TAVI will replace open AVR for the management of uncomplicated, severe AS.

## **5. Future directions**

The evidence base for TAVI is rapidly evolving, and there has been a significant rise in the number of new publications over the last five years. Most eagerly awaited are the findings of Cohort A of the PARTNER Trial. This will provide insight into the comparative efficacy of TAVI and open AVR in high-risk patients with severe symptomatic AS. The results of this trial will have important implications for healthcare policy implementation and may mean greater financial provision for TAVI in high-risk patients. In addition, the two main techniques for accessing the aortic valve (transfemoral and transapical) will need to be compared in an RCT, especially given that pooled data suggests one-year survival is worse after transapical TAVI (Figulla *et al.*, 2011). If this is proved to be the case, then the transapical approach may be restricted to patients in whom the transfemoral route is contraindicated.

Transcatheter Aortic Valve Implantation 243

Bagur, R., Webb, JG., Nietlispach, F., *et al.* (2010). Acute kidney injury following

comparison with surgical aortic valve replacement, *Eur Heart J* 31: 865–874. Behan, M., Haworth, P., Hutchinson, N., Trivedi, U., Laborde, JC., & Hildick-Smith, D

Berry, C., Asgar, A., Lamarche, Y., *et al.* (2007). Novel therapeutic aspects of percutaneous

Bonow, R. W., Carabello, B. A., Kanu, C., *et al.* (2006). ACC/AHA 2006 guidelines for the

Brown, M. L., Schaff, H. V., Sarano, M. E., *et al.* (2008). Is the European System for cardiac

Brown, J. M., O'Brien, S. M., Wu, C., Sikora, J. A., Griffith, B. P., & Gammie, J. S. (2009).

Charlson, E., Legedza, A. T. R., & Hamel, M. B. (2006). Decision-making and outcomes in

Cheitlin, M. D., Gertz, E. W., Brundage, B. H., Carlson, C. J., Quash, J. A., & Bode, R. S. Jr

Clavel, M. A., Webb, J. G., Rodés-Cabau, J., *et al.* (2010). Comparison between transcatheter

Connolly, H. M., Oh, J. K., Schaff, H. V., *et al.* (2000). Severe aortic stenosis with low

Cribier, A., Savin, T., Saoudi, N., Rocha, P., Berland, J., & Letac, B (1986). Percutaneous

Cribier, A., Eltchaninoff, H., Bash, A., *et al.* (2002). Percutaneous transcatheter implantation

Davies, S. W., Gershlick, A. H., & Balcon, R. (1991). Progression of valvular aortic stenosis: a

De Jaegere, P., Van Dijk, L. C., Laborde, J. C., *et al.* (2007). True percutaneous implantation of

and reduced left ventricular ejection fraction, *Circulation* 122: 1928-1936. Clavel, MA., Webb, JG., Pibarot, P., e*t al.* (2009). Comparison of the hemodynamic

Surgeons National Database, *J Thorac Cardiovasc Surg* 137: 82-90.

severe symptomatic aortic stenosis, *J Heart Valve Dis* 15: 312-321.

aortic stenosis, *J Am Coll Cardiol* 53: 1883–1891.

replacement in 52 patients, *Circulation* 101: 1940-1946.

alternative to valve replacement? *Lancet* 1: 63-67.

long-term retrospective study, *Eur Heart J* 12: 10-14.

*Catheter Cardiovasc Interv* 72: 1012-1015.

Guidelines, *Circulation* 114: e84-e231.

*Cardiovasc Interv* 70: 610-616.

566-571.

*Heart J* 98: 689-700.

*Circulation* 106: 3006-3008.

transcatheter aortic valve implantation: predictive factors, prognostic value, and

(2008). Percutaneous aortic valve implants under sedation: our initial experience,

aortic valve replacement with the 21F CoreValve Revalving system, *Catheter* 

management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice

Operative Risk Evaluation model valid for estimating operative risk of patients considered for percutaneous aortic valve replacement? *J Thorac Cardiovasc Surg* 136:

Isolated aortic valve replacement in North America comprising 108,687 patients in 10 years: changes in risks, valve types, and outcomes in the Society of Thoracic

(1979). Rate of progression of severity of valvular aortic stenosis in the adult, *Am* 

and surgical prosthetic valve implantation in patients with severe aortic stenosis

performance of percutaneous and surgical bioprostheses for the treatment of severe

transvalvular gradient and severe left ventricular dysfunction: result of aortic valve

transluminal valvuloplasty of acquired aortic stenosis in elderly patients: an

of an aortic valve prosthesis for calcific aortic stenosis: first human case description,

the CoreValve aortic valve prosthesis by the combined use of ultrasound guided vascular access, Prostar XL and the TandemHeart, *EuroIntervention* 2: 500–505.

Using TAVI in moderate- or low-risk patients is not currently justified, because it would be unethical to withhold access to open AVR with its proven efficacy. The results of long-term follow-ups (> 10 years) of high-risk patients will be necessary to assess the durability of TAVI implants and to inform decisions about their use in younger, fitter patients. Given the short life expectancy of patients denied open surgery, it is unlikely that sufficient data will be available in the foreseeable future to determine whether TAVI is appropriate in low-risk candidates. It is also worth noting that a number of new prostheses will undoubtedly emerge in the next decade and that they will require appropriate evaluation against existing gold standards. An RCT using the Medtronic CoreValve system is also anticipated to publish its findings in 2013. Once the feasibility, safety, efficacy, and durability of TAVI devices have been established, the onus will shift towards healthcare economic evaluation to identify the most cost-effective means of treating severe AS. It is certainly conceivable that minimally invasive techniques, such as TAVI, will prove cost-effective in the long-term.

### **6. Conclusions**

Severe aortic stenosis has a poor prognosis once symptoms of congestive cardiac failure and angina develop. It is conventional wisdom that surgical AVR offers the best hope of symptom palliation and long-term survival. However, approximately one-third of patients with severe symptomatic AS are denied access to surgery because of comorbidities and high operative risk. The last decade has witnessed the introduction of transcatheter aortic valve implantation, a surgical innovation that permits percutaneous replacement of the diseased aortic valve without sternotomy and a cardiopulmonary bypass. TAVI remains in its infancy, yet it has demonstrated superior medium-term survival, fewer symptoms, and a better quality of life than medical care alone. Doubts persist over the durability of the implants, long-terms outcomes, and the relative efficacy compared with surgical AVR in high-risk patients. However, TAVI is set to continue to revolutionize the management of severe symptomatic aortic stenosis.

#### **7. References**


Using TAVI in moderate- or low-risk patients is not currently justified, because it would be unethical to withhold access to open AVR with its proven efficacy. The results of long-term follow-ups (> 10 years) of high-risk patients will be necessary to assess the durability of TAVI implants and to inform decisions about their use in younger, fitter patients. Given the short life expectancy of patients denied open surgery, it is unlikely that sufficient data will be available in the foreseeable future to determine whether TAVI is appropriate in low-risk candidates. It is also worth noting that a number of new prostheses will undoubtedly emerge in the next decade and that they will require appropriate evaluation against existing gold standards. An RCT using the Medtronic CoreValve system is also anticipated to publish its findings in 2013. Once the feasibility, safety, efficacy, and durability of TAVI devices have been established, the onus will shift towards healthcare economic evaluation to identify the most cost-effective means of treating severe AS. It is certainly conceivable that minimally invasive techniques, such as TAVI, will prove cost-effective in the long-term.

Severe aortic stenosis has a poor prognosis once symptoms of congestive cardiac failure and angina develop. It is conventional wisdom that surgical AVR offers the best hope of symptom palliation and long-term survival. However, approximately one-third of patients with severe symptomatic AS are denied access to surgery because of comorbidities and high operative risk. The last decade has witnessed the introduction of transcatheter aortic valve implantation, a surgical innovation that permits percutaneous replacement of the diseased aortic valve without sternotomy and a cardiopulmonary bypass. TAVI remains in its infancy, yet it has demonstrated superior medium-term survival, fewer symptoms, and a better quality of life than medical care alone. Doubts persist over the durability of the implants, long-terms outcomes, and the relative efficacy compared with surgical AVR in high-risk patients. However, TAVI is set to continue to revolutionize the management of

Al-Attar, N., Himbert, D., Descoutures, F., *et al.* (2009). Transcatheter aortic valve

Ambler, G., Omar, R. Z., Royston, P., Kinsman, R., Keogh, B. E., & Taylor, K. M. (2005).

Anderson, H. R., Knudsen, L. L., & Hasenkam, J. M. (1992). Transluminal implantation of

Anselmi, A., Abbate, A., Girola, F., *et al.* (2004). Myocardial ischaemia, stunning,

Asimakopoulos, G., Edwards, M. B., & Taylor, K. M. (1997). Aortic valve replacement in

implantation: selection strategy is crucial for outcome, *Ann Thorac Surg* 87(6): 1757–

Generic, simple risk stratification model for heart valve surgery, *Circulation* 112:

artificial heart valves. Description of a new expandable aortic valve and initial results with implantation by catheter technique in closed chest pigs, *Eur Heart J* 13:

inflammation, and apoptosis during cardiac surgery: a review of evidence, *Eur J* 

patients 80 years of age and older: survival and cause of death based on 1100 cases: collective results from the UK Heart Valve Registry, *Circulation* 96: 3403-3408.

**6. Conclusions** 

**7. References** 

1762.

224–231.

704-708.

*Cardiothorac Surg* 25: 304-311.

severe symptomatic aortic stenosis.


Transcatheter Aortic Valve Implantation 245

Gotzmann, M., Hehen, T., Germing, A., *et al.* (2010). Short-term effects of transcatheter aortic

Grube, E., Laborde, J. C., Gerckens, U., *et al.* (2006). Percutaneous implantation of the

Grube, E., Schuler, G., Buellesfeld, L., *et al.* (2007). Percutaneous aortic valve replacement for

Grube, E., Buellesfeld, L., Mueller, R., *et al.* (2008). Progress and current status of

Gurvitch, R., Wood, D. A., Tay, E. L., *et al.* (2010). Transcatheter aortic valve implantation:

Gurvitch, R., Cheung, A., Bedogni, F., & Webb, J. G. (2011). Coronary obstruction following

Hammermeister, K. E., Sethi, G. K., Henderson, W. G., Oprian, C., Kim, T., & Rahimtoola, S.

Hattler, B. G., Madia, C., Johnson, C., *et al.* (1994). Risk stratification using the Society of

Higgins, T. L., Estafanous, F. G., Loop, F. D., Beck, G. J., Blum, J. M., & Paranandi, L. (1992).

Himbert, D., Descoutures, F., Al-Attar, N., *et al.* (2009). Results of transfemoral or transapical

Iung, B., Baron, G., Butchart, E. G., *et al.* (2003). A prospective survey of patients with

Iung, B., Cachier, A., Baron, G., *et al.* (2005). Decision-making in elderly patients with severe aortic stenosis: why are so many denied surgery? *Eur Heart J* 26: 2714-2720. Jamieson, W. R., Burr, L. H., Tyers, G. F., *et al.* (1995). Carpentier-Edwards supraannular

Jilaihawi, H., Chin, D., Spyt, T., *et al.* (2010). Prosthesis-patient mismatch after transcatheter

Kahlert, P., Knipp, S. C., Schlamann, M., *et al.* (2010). Silent and apparent cerebral ischemia

magnetic resonance imaging study, *Circulation* 121: 870–878.

test in severe and symptomatic aortic stenosis, *Heart* 96: 1102-1106.

disease: the Siegburg first-in-man study, Circulation 114: 1616-24.

CoreValve Revalving System, *Circ Cardiovasc Interv* 1(3): 167–175.

with a mechanical or bioprosthesis, *N Engl J Med* 328: 1289-1296.

coronary artery bypass patients, *J Am Med Assoc* 267: 2344-2348.

Thoracic Surgeons program, *Ann Thorac Surg* 52: 1348-1352.

aortic stenosis, *J Am Coll Cardiol* 54(4): 303–311.

Disease, *Eur Heart J* 24: 1231-1243.

S235–S240.

31:857–864.

outcome, *J Am Coll Cardiol* 50: 69-76.

cohort, *Circulation* 122: 1319–1327.

*Catheter Cardiovasc Interv* 77: 439-444.

valve implantation on neurohormonal activation, quality of life and 6-minute walk

CoreValve self-expanding valve prosthesis in high-risk patients with aortic valve

severe aortic stenosis in high-risk patients using the second- and current thirdgeneration self expanding CoreValve prosthesis. Device success and 30-day clinical

percutaneous aortic valve replacement: results of three device generations of the

durability of clinical and hemodynamic outcomes beyond 3 years in a large patient

transcatheter aortic valve-in-valve implantation for failed surgical bioprostheses,

(1993). A comparison of outcomes in men 11 years after heart-valve replacement

Stratification of morbidity and mortality outcome by preoperative risk factors in

aortic valve implantation following a uniform assessment in high-risk patients with

valvular heart disease in Europe: The Euro Heart Survery on Valvular Heart

porcine bioprosthesis: clinical performance to twelve years, *Ann Thorac Surg* 60:

aortic valve implantation with the Medtronic-Corevalve bioprosthesis, *Eur Heart J*

after percutaneous transfemoral aortic valve implantation: a diffusion-weighted


De Jaegere, P. P., Piazza, N., Galema, T. W., *et al.* (2008). Early echocardiographic evaluation

Demers, C., McKelvie, RS., Negassa, A., & Yusuf, S. (2001). Reliability, validity, and

Descoutures, F., Himbert, D., Lepage, L., *et al.* (2008). Contemporary surgical or

Dewey, T. M., Brown, D., Ryan, W. H., Herbert, M. A., Prince, S. L., & Mack, M. J. (2008).

Edwards, F. H., Peterson, E. D., Coombs, L. P., *et al.* (2001). Prediction of operative mortality

Elayda, MA., Hall, RJ., Reul, RM., *et al.* (1993). Aortic valve replacement in patients 80 years and older, operative risks and long-term results, *Circulation* 88: 11-16. Eltchaninoff, H., Prat, A., Gilard, M., *et al.* (2011). Transcatheter aortic valve implantation:

Falk, V., Schwammenthal, E. E., Kempfert, J., *et al.* (2009). New anatomically oriented

Fayers, P., & Hays, R. (2005). Assessing quality of life in clinical trials. 2nd ed. Oxford:

Ferrari, E., Marcucci, C., Sulzer, C., & von Segesser, LK (2010). Which available transapical

Figulla, L., Neumann, A., Figulla, H. R., Kahlert, P., Erbel, R., & Neumann, T (2011).

Florath, I., Rosendahl, U. P., Mortasawi, A., *et al.* (2003). Current determinants of operative

Florath, I., Albert, A., Boening, A., Ennker, I. C., & Ennker, J., (2010). Aortic valve

Gabrielle, F., Roques, F., Michel, P., *et al.* (1997). Is the Parsonnet's score a good predictive

Ghanem, A., Muller, A., Nahle, C. P., *et al.* (2010). Risk and fate of cerebral embolism after

weighted magnetic resonance imaging, *J Am Coll Cardiol* 55: 1427–1432. Gogas, B. D., Zacharoulis, A. A., & Antoniadis, A. G. (2011). Acute coronary occlusion

transapical aortic valve implantation, *Ann Thorac Surg* 87: 925-926.

after valve replacement surgery, *J Am Coll Cardiol* 37: 885–892.

System aortic valve bioprosthesis, *Eurointervention* 4: 351-357.

142: 698-703.

1410-1417.

135: 180-187.

registry, *Eur Heart J* 32(2): 191–197.

Oxford University Press.

*Surg* 11: 83-85.

100: 265-276.

*Surg* 37: 1304-1310.

study, Eur J Cardiothoracic Surg 11: 400–411.

following TAVI, *Catheter Cardiovasc Interv* 77: 435-438.

75-83.

following percutaneous implantation with the self-expanding CoreValve Revalving

responsiveness of the six- minute walk test in patients with heart failure, *Am Heart J*

percutaneous management of severe aortic stenosis in the elderly, *Eur Heart J* 29:

Reliability of risk algorithms in predicting early and late operative outcomes in high-risk patients undergoing aortic valve replacement, *J Thorac Cardiovasc Surg*

early results of the FRANCE (FRench Aortic National CoreValve and Edwards)

transcatheter valve fits into degenerated bioprostheses? *Interact Cardiovasc Thorac* 

Transcatheter aortic valve implantation: evidence on safety and efficacy compared with medical therapy. A systematic review of current literature, *Clin Res Cardiol*

mortality in 1400 patients requiring aortic valve replacement, *Ann Thorac Surg* 76:

replacement in octogenarians: identification of high-risk patients, *Eur J Cardiothorac* 

score of mortality in adult cardiac surgery? Assessment by a French multicentre

transfemoral aortic valve implantation: a prospective pilot study with diffusion-


Transcatheter Aortic Valve Implantation 247

Moss, R. R., Ivens, E., Pasupati, S., *et al.* (2008). Role of echocardiography in percutaneous

Myken, P. S. U., Caidahl, K., Larsson, P., Larsson, S., Wallentin, I., & Berggren, H. E. (1995).

Nashef, S. A. M., Roques, F., Michel, P., Gauducheau, E., Lemeshow, S., & Salamon, R.

Nietlispach, F., Wijesinghe, N., Gurvitch, R., *et al.* (2010). An embolic deflection device for

Nissinen, J., Biancari, F., Wistbacka, J. O., *et al.* (2009). Is it possible to improve the accuracy

Nowicki, ER., Birkmeyer, NJO., Weintraub, RW., *et al.* (2004). Multivariable prediction of in-

O'Brien, SM., Shahian, DM., Filardo, G., *et al.* (2009). The Society of Thoracic Surgeons 2008 cardiac risk models: part 2 – isolated valve surgery, *Ann Thorac Surg* 88: S23-S42. Olsen, L. K., Engstrøm, T., & Søndergaard, L (2010). Transcatheter valve-in-valve

Otten, A. M., van Domburg, R. T., van Gameren, M., *et al.* (2008). Population characteristics,

Otto, C. M., Pearlman, A. S., & Gardner, C. L. (1989). Hemodynamic progression of aortic

Otto, C. M., Mickel, M. C., Kennedy, J. W., *et al.* (1994). Three-year outcome after balloon

Parsonnet, V., Dean, D., & Bernstein, A. D. (1989). A method of uniform stratification of risk

Peter M, Hoffmann A, Parker C, Lüscher T, Burckhardt D (1993). Progression of aortic stenosis: role of age and concomitant coronary artery disease, *Chest* 103: 1715-1719. Peterseim, D. S., Cen, Y. Y., Cheruvu, S., *et al.* (1999). Long-term outcome after biologic

Piazza, N., Grube, E., Gerckens, U., *et al.* (2008). Procedural and 30-day outcomes following

Piazza, N., Nuis, R. J., Tzikas, A., *et al.* (2010). Persistent conduction abnormalities and

following CE mark approval, *Eurointervention* 4: 242-249.

implantation. *EuroIntervention*, 6, pp. 475–484.

percutaneous valve replacement, *EuroIntervention* 4(2): 250–255.

Mechanical versus biological valve prosthesis: a ten-year comparison regarding

(1999). European system for cardiac operative risk evaluation (EuroSCORE), *Eur J* 

hospital mortality associated with aortic and mitral valve surgery in northern New

implantation due to severe aortic regurgitation in a degenerated aortic homograft, J

treatment assignment and survival of patients with aortic stenosis referred for

stenosis in adults assessed by Doppler echocardiography, *J Am Coll Cardiol* 13: 545-

aortic valvuloplasty. Insights into prognosis of valvular aortic stenosis, *Circulation*

for evaluating the results of surgery in acquired adult heart disease, *Circulation* 79:

versus mechanical aortic valve replacement in 841 patients, *J Thorac Cardiovasc Surg*

transcatheter aortic valve implantation using the third generation (18 Fr) CoreValve revalving system: results from the multicentre, expanded evaluation registry 1-year

requirements for pacemaking six months after transcatheter aortic valve

aortic valve implantation, *J Am Coll Cardiol Imaging* 1:15–24.

function and quality of life, *Ann Thorac Surg* 60: S447-S452.

aortic valve interventions, *JACC Cardiovasc Interv* 3: 1133–1138.

of the EuroSCORE? *Eur J Cardiothorac Surg* 36: 799-804.

England, *Ann Thorac Surg* 77: 1966-1977.

Invasive Cardiol 21:E197-E200.

550.

89: 642-650.

117: 890-897.

I3-I12.

*Cardiothorac Surg* 16: 9-13.


Kapadia, S. R., Goel, S. S., Svensson, L. G., *et al.* (2009). Characterization and outcome of

Kelly, T. A., Rothbart, R. M., Cooper, C. M., Kaiser, D. L., Smucker, M. L., & Gibson, R. S.

than 20 years of age with valvular aortic stenosis, *Am J Cardiol* 61: 123-130. Kempfert, J., Van Linden, A., Linke, A., *et al.* (2010). Transapical off-pump valve-in-valve

Khawaja, M. Z., Haworth, P., Ghuran, A., *et al.* (2010). Transcatheter aortic valve

failed bioprosthetic aortic valve replacements, *J Am Coll Cardiol* 55: 97-101. Kobayashi, K. J., Williams, J. A., Nwakanma, L. U., *et al.* (2009). EuroSCORE predicts short-

Krane, M., Deutsch, M. A., Bleizffer, S., *et al.* (2010). Quality of life among patients undergoing transcatheter aortic valve implantation, *Am Heart J* 160: 451-457. Latsios, G., Gerckens, U., & Grube, E (2010). Transaortic transcatheter aortic valve

Lee, S. J., Lindquist, K., Segal, M. R., & Covinsky, K. E (2006). Development and validation of a prognostic index for 4-year mortality in older adults, *JAMA* 295 : 801-808. Lefevre, T., Kappetein, A. P., Wolner, E., *et al.* (2011). One year follow-up of the multi-centre European PARTNER transcatheter heart valve study, *Eur Heart J* 32(2):148–157. Leon, M. B., Smith, C. R., Mack, M., *et al.* for the PARTNER Trial Investigators (2010).

Leontyev, S., Walther, T., Borger, MA., *et al.* (2009). Aortic valve replacement in

Lieberman, E. B., Bashore, T. M., Hermiller, J. B., *et al.* (1995). Balloon aortic valvuloplasty in

Low, R. I., Bolling, S. F., Yeo, K. K., & Ebner, A. (2008). Direct flow medical percutaneous

Malaisrie, S. C., McCarthy, P. M., McGee, E. C., *et al.* (2010). Contemporary perioperative

Malaisrie, S. C., Tuday, E., Lapin, B., *et al.* (2011). Transcatheter valve implantation decreases the rate of unoperated aortic stenosis, *Eur J Cardiothorac Surg* 40(1): 43-48. Melby, S. J., Zierer, A., Kaiser, S. P., *et al.* (2007). Aortic valve replacement in octogenarians: risk factors for early and late mortality, *Ann Thorc Surg* 83: 1651-1657. Monin, J. L., Quere, J. P., Monchi, M., *et al.* (2003). Low-gradient aortic stenosis: operative

aortic valve: Proof of concept, *EuroInterv* 4: 256–261.

dobutamine stress hemodynamics, *Circulation* 108: 319-324.

valve replacement, *J Thorac Cardiovasc Surg* 137(6): 1430– 1435.

1934-1941.

bypass patients, *J Card Surg* 24: 637-643.

*Cardiovasc Interv* 75: 1129-1136.

*Eng J Med* 363: 1578-1607.

1440-1445.

1522-1528.

751-756.

patients with severe symptomatic aortic stenosis referred for percutaneous aortic

(1988). Comparison of outcome of asymptomatic to symptomatic patients older

implantation in patients with degenerated aortic xenografts, *Ann Thorac Surg* 89:

implantation for stenosed and regurgitant aortic valve bioprostheses CoreValve for

and mid-term mortality in combined aortic valve replacement and coronary artery

implantation: a novel approach for the truly "no-access option" patients, *Catheter* 

Transcatheter aortic-valve implantation in patients who cannot undergo surgery, *N* 

octogenarians: utility of risk stratification with EuroSCORE, *Ann Thorac Surg* 87:

adults: failure of procedure to improve long-term survival, *J Am Coll Cardiol* 26:

results of isolated aortic valve replacement for aortic stenosis, *Ann Thorac Surg* 89:

risk stratification and predictors for long-term outcome: a multicenter study using


Transcatheter Aortic Valve Implantation 249

Tamburino, C., Capodanno, D., Mul, E. M., *et al.* (2009). Procedural success and 30-day

Thielmann, M., Wendt, D., Eggebrecht, H., *et al.* (2009). Transcatheter aortic valve

Thomas, M., Schymik, G., Walther, T., *et al.* (2010). Thirty-day results of the SAPIEN aortic

Tops, L. F., Wood, D. A., Delgado, V., *et al.* (2008). Noninvasive evaluation of the aortic root

Treede, H., Tubler, T., Reichenspurner, H., *et al.* (2010). Six-month results of a repositionable

Tu, J. V., Jaglal, S. B., & Naylor, C. D. (1995). Multicenter validation of a risk index for

Turina, J., Hess, O., Sepulcri, F., & Krayenbuehl, H. P. (1987). Spontaneous course of aortic

Ussia, G. P., Mule, M., Barbanti, M., *et al.* (2009). Quality of life assessment after

Ussia, G. P., Capodanno, D., Brbanti, M., *et al.* (2010). Balloon aortic valvuloplasty for severe

Vahanian, A., Baumgartner, H., Bax, J., *et al.* (2007). Task Force on the Management of

Vahanian, A., Alfieri, O. R., Al-Attar, N., *et al.* (2008). Transcatheter valve implantation for

Vahanian, A., Himbert, D., & Brochet, E. (2010). Transcatheter valve implantation for

Vavuranakis, M., Voudris, V., Vrachatic, D. A., *et al.* (2010). Transcatheter aortc valve

the intervention without general anaesthesia, *Hellenic J Cardiol* 51: 492-500.

direct flow medical aortic valve, *J Thorac Cardiovasc Surg* 140: 897-903. Tremblay, N. A., Hardy, J. F., Perault, J., & Carrier, M. (1993). A simple classification of the

risk in cardiac surgery: the first decade, *Can J Anaesth* 40: 103-111.

percutaneous aortic valve implantation, *Eur Heart J* 30: 1790-1796.

*Catheter Cardiovasc Interv* 75: 51- 55.

*Circulation* 122: 62–69.

surgery, *Circulation* 91: 677-684.

*Invasive Cardiol* 22: 161-166.

valve disease, *Eur Heart J* 8: 471-483.

of Cardiology, *Eur Heart J* 28: 230-268.

Interventions (EAPCI), *Eur Heart J* 29: 1463-1470.

patients with severe aortic stenosis, *Heart* 96: 1849-1856.

replacement, *Ann Thorac Surg* 88(5): 1468– 1474.

replacement, *J Am Coll Cardiol Imaging* 1: 321-330.

clinical outcomes after percutaneous aortic valve replacement using current thirdgeneration self-expanding CoreValve prosthesis. *J Invasive Cardiol*, 21: 99-100. Taramasso, M., Sharp, A. S., & Maisano, F. (2010). First-in-man case report of the use of an

Edwards Sapien valve to treat regurgitant CoreValve aortic valve prosthesis,

implantation in patients with very high risk for conventional aortic valve

bioprosthesis European outcome (SOURCE) registry: a European registry of transcatheter aortic valve implantation using the Edwards SAPIEN valve,

with multislice computed tomography: implications for transcatheter aortic valve

and retrievable pericardial valve for transcatheter aortic valve replacement: The

mortality, intensive care unit stay, and overall hospital length of stay after cardiac

aortic stenosis as a bridge to high-risk transcatheter aortic valve implantation, *J* 

Valvular Heart Disease of the European Society of Cardiology; ESC Committee for Practice Guidelines. Guidelines on the management of valvular heart disease: The Task Force on the Management of Valvular Heart Disease of the European Society

patients with aortic stenosis: a position statement from the European association of Cardio-Thoracic Surgery (EACTS) and the European Society for Cardiology (ESC) in collaboration with the European Association of Percutaneous Cardiovascular

implantation, patient selection process and procedure: two centres' experience of


Pons, J. M. V., Granados, A., Espinas, J. A., Borras, J. M., Martin, I., & Moreno, V. (1997).

Rajani, R., Buxton, W., Haworth, P., *et al.* (2010). Prognostic benefit of transcatheter aortic

Reynolds, M. R., Moses, J. W., Svensson, L. G., *et al.* (2008). For the REVIVAL II

Rodes-Cabau, J., Dumont, E., De La Rochelliere, R., *et al.* (2008). Feasibility and initial results

Rodes-Cabau, J., Dumont, E., & Doyle, D. (2009). "Valve-in-valve" for the treatment of

Rodes-Cabau, J., Webb, J. G., Cheung, A., *et al.* (2010). Transcatheter aortic valve

Roques, F., Gabrielle, F., Michel, P., de Vincentiis, C., David, M., & Baudet, E. (1995). Quality

Roques, F., Nashef, S. A., Michel, P., *et al.* (1999). Risk factors and outcome in European

Schofer, J., Schluter, M., Treede, H., *et al.* (2008). Retrograde transarterial implantation of a

Sharp, A. S., Michex, I., & Colombo, A. (2010). First transaxillary implantation of Edwards

Sherif, MA., Abdel-Wahab, M., Stocker, B., *et al.* (2010). Anatomic and procedural predictors

Spargias, K., Manginas, A., Pavlides, G., Khoury, M., Stavridis, G., & Rellia, P. (2008).

Svensson, L. G., Dewey, T., Kapadia, S., *et al.* (2008). United States feasibility study of

risk model, *Eur J Cardiothorac Surg* 11: 415-423.

B122.

1246.

133.

75: 507-510.

397-407.

*Thorac Surg* 86: 46-55.

*Cardiovasc Interv* 74: 1116-1119.

Canadian experience, *J Am Coll Cardiol* 50: 69-76.

patients, *Eur J Cardiothorac Surg* 15: 816-823.

bioprosthesis, *J Am Coll Cardiol* 56: 1623–1629.

French multicenter study, *Eur J Cardiothorac Surg* 9: 433-440.

Ross, J. Jr., & Braunwald, E. (1968). Aortic stenosis, *Circulation* 38: 61-67.

aortic stenosis, *Catheter Cardiovasc Interv* 75(7): 1121–1126.

Assessing open heart surgery mortality in Catalonia (Spain) through a predictive

valve implantation compared with medical therapy in patients with inoperable

Investigators. Quality of life improvements following transcatheter aortic valve replacement: results from the REVIVAL II feasibility study, *J Am Coll Cardiol* 51: B1-

of percutaneous aortic valve implantation including selection of the transfemoral or transapical approach in patients with severe aortic stenosis, *Am J Cardiol* 102 : 1240-

paravalvular leaks fol- lowing transcatheter aortic valve implantation, *Catheter* 

implantation for the treatment of severe symptomatic aortic stenosis in patients at very high or prohibitive surgical risk acute and late outcomes of the multicenter

of care in adult heart surgery: proposal for a self-assessment approach based on a

cardiac surgery: an analysis of the EuroSCORE multinational database of 19030

nonmetallic aortic valve prosthesis in high-surgical-risk patients with severe aortic stenosis. A first-in-man feasibility and safety study, *Circ Cardiovasc Intervent* 1:126-

Sapien valve to treat incompetent aortic bioprosthesis, Catheter Cardiovasc Interv

of paravalvular aortic regurgitation after implantation of the Medtronic CoreValve

Transcatheter aortic valve implantation: first Greek experience, *Hellenic J Cardiol* 49:

transcatheter insertion of a stented aortic valve by the left ventricular apex, *Ann* 


**13** 

*Germany* 

**Image-Guided Transcatheter Aortic** 

Mohamed Esmail Karar1, David Holzhey2,

Friedrich-Wilhelm Mohr2 and Oliver Burgert1

Matthias John3, Ardawan Rastan2,

*3Siemens AG, Healthcare Sector, Forchheim* 

**Valve Implantation Assistance System** 

*2Department of Cardiac Surgery, Heart Center, University of Leipzig* 

*1Innovation Center Computer Assisted Surgery (ICCAS), University of Leipzig* 

Transcatheter aortic valve implantation (TAVI) is a recently developed surgical technique to treat symptomatic aortic valve stenosis in elderly and high-risk patients (Eltchaninoff et al., 2008; Ferrari & von Segesser, 2010). Compared to the standard aortic valve replacement surgery, the TAVI limits the surgical access to either a small minithoracotomy (transapical TAVI) or femoral approach (transfemoral TAVI) causing minimal tissue trauma. Independently of the TAVI approach it can be performed on the beating heart without cardiopulmonary bypass support (Walther et al., 2009). Recovery time may be reduced and the patient can eventually return to normal activity more quickly. More than 70,000 transcatheter valve implantations have been performed worldwide (Valle-Fernández et al.,

The TAVI is done via a retrograde (transfemoral, transaxillary) or antegrade (transapical) approach (Singh et al., 2008). The main advantage of the transapical TAVI technique is the direct access to the aortic valve which eliminates the need for a large peripheral vascular access in patients with peripheral vascular disease, small tortuous vasculature, a history of

In transapical TAVI (Walther et al., 2009), a stented valve bioprosthesis that is temporarily crimped upon a balloon catheter, is inserted through the apex into the aortic root via a left anterolateral minithoracotomy. For that the apex of the left ventricle is punctured with a needle, and after balloon valvuloplasty the aortic valve prosthesis (AVP) is positioned within the stenotic aortic valve using guide wire techniques. After reaching the correct position, the stented AVP is deployed by an inflatable balloon to reach its final diameter,

The Edwards SAPIENTM prosthesis (Edwards Lifesciences Inc, Irvine, CA, USA) is the most commonly used prosthesis for TAVI in several European countries and the only one approved for transapical approach so far (Thomas et al., 2010). Thus the Edwards SAPIENTM prosthesis has been used in this study. It consists of three bovine pericardial cusps mounted

major vascular complications, or previous vascular interventions (Singh et al., 2008).

thus fixing the prosthesis to the aortic annulus (Fig. 1a).

into a stainless-steel balloon-expandable stent (Fig. 1b).

**1. Introduction** 

2010).

