Edwards Sapien Aortic Valve: Transfemoral Approach

VASCULAR DISEASE MANAGEMENT  2010;7(1):E1-E9

Abstract

The development of transcatheter aortic valve implantation may become the primary therapy for high-risk patients with severe symptomatic aortic stenosis. The Edwards Sapien valve (Edwards Lifesciences, Irvine, California) consists of a balloon-expandable stent with an integrated bovine pericardial valve. The technique and the device itself have rapidly evolved; more than 5,000 patients worldwide have undergone implantation with this valve. This review will focus on the Edwards Sapien balloon-expandable heart valve using the transfemoral approach. Risk assessment is important since the current indication is restricted to high surgical risk or non-operable patients, while appropriate patient selection and screening is crucial for success and for avoiding complications. Careful evaluation of the iliofemoral vessels by angiography and high-quality computed tomography (CT) is indispensable since vascular injury is the most common cause of morbidity andmortality for these procedures. Precise measurement of the aortic annulus diameter for valve sizing is very important when choosing the appropriate valve size. The relation among the annulus, plaque in the left coronary leaflet, and the distance to the left coronary ostium is also important. Existing follow-up data show no evidence of restenosis or prosthetic valve dysfunction, and randomized trials currently underway will confirm procedural safety and guide the applicability of this technology.

With a prevalence of 4.6% in adults over 75 years of age, aortic stenosis (AS) is the most common valvular heart disease in an aging population.¹ Surgical aortic valve replacement is the recommended treatment for symptomatic patients. However, a large proportion of patients are not referred or are deferred due to high risk for surgery.² Data from the Euro Heart Survey on valvular heart disease revealed that up to 30% of patients with severe AS do not undergo surgery. Trans-catheter aortic valve implantation (TAVI) has been designed to address a new treatment option for this high-risk population. The first human implantation of a percutaneous aortic valve was performed by Alain Cribier in April 2002 with a balloon-expandable valve on a stent. Since then, this procedure has rapidly expanded, and as of today, more than 5,000 patients have had the aortic valve implanted. The Edwards Sapien valve (Edwards Lifesciences, Irvine, California), available in 23 mm and 26 mm, consists of a balloon-expandable stainless steel frame with an integrated trileaflet bovine pericardial valve. The latest version of this valve has a chromium cobalt frame. Valve implantation can be performed safely and effectively utilizing both the transfemoral approach and transapical placement. This article will review the present state of TAVI using the Edwards Sapien valve via the transfemoral approach.

Patient Selection and Risk Assessment

TAVI is intended for use in symptomatic patients with severe calcific AS requiring aortic valve replacement who are at high risk for open chest surgery due to comorbid conditions, and for patients who are inoperable. Defining high-risk surgical patients is not simple. A Society of Thoracic Surgeons (STS) risk score > 10 and/or a logistic EuroSCORE > 20 are most often used to define high risk. The EuroSCORE has been shown to predict long-term mortality after valve surgery.³ This algorithm, however, has been shown to persistently overestimate the mortality rate,⁴ and this overestimation is greatest in high-risk patients.5 The STS score has been shown to underestimate the true morality rate after cardiac surgery, but it more closely reflects the operative- and 30-day mortality for the highest-risk patients undergoing aortic valve replacement.⁶ Patients can be at very high operative mortality risk, yet have low scores. There are numerous comorbidities not captured in the EuroSCORE and STS scoring systems such as porcelain aorta, chest-wall radiation, chest-wall deformity, highly compromised respiratory function, frailty, cirrhosis, and others. There have been attempts to quantify the frailty index⁷ and correlate the frailty index with outcomes,⁸ however, frailty analysis is particularly difficult and is often not quantifiable. The clinical judgment of experienced cardiac surgeons plays a key role in assessing operative mortality in these cases. Currently available validated risk score systems have not captured the “non-operable” patient. The definition of “inoperability” is difficult, and often requires the consensus of several surgeons. It is important to emphasize that those scoring systems are not intended to be used as substitutes for clinical decision making.

This new treatment for the high-risk populations is aimed at those patients whose comorbidities will not interfere with normal recovery after aortic valve implantation. For example, patients who are bedridden or have a life expectancy of Screening Adequate vascular access is one of the most important determinants of procedural success and/or complications. Three imaging modalities are available to evaluate the access vessels: angiographic diameters measured with quantitive coronary angiography (QCA) and a reference marker pigtail (Figure1), contrast computed tomography (CT), and intravascular ultrasound (IVUS). CT is the most accurate modality and provides the most useful information to predict the feasibility of a transfemoral approach. We have developed a novel technique for high-quality, low-contrast CT of the aorta and iliofemoral vessels. After diagnostic catheterization, a 4 French (Fr) pigtail catheter is placed in the abdominal aorta just below the renal arteries. In the CT room, the power injector is connected to the pigtail catheter and 10–15 cc of contrast mixed with normal saline in a 1:3 to 1:4 dilution is injected at 4 cc per second. A helical CT of the abdomen and pelvis is acquired (64/256 x 0.625 mm collimation, rotation time 0.75 sec, pitch 0.64, 120 Kv 154 mAs). This technique allows the practitioner to use only a minimum amount of contrast and helps preserve renal function in this population of older patients with critical AS who often have renal dysfunction⁹ (Figures 2A and B). Precise vessel measurements are performed in multiple sites in the common femoral, external, and common iliac arteries. The vessel diameter is carefully assessed in longitudinal and axial views. The cursor is placed in the abdominal aorta and is then brought down millimeter by millimeter to get exact measurements throughout the entire length of the vessel considered for access. The 22 Fr (outer diameter 25.5 Fr = 8.5 mm) and 24 Fr (outer diameter 27.9 Fr = 9.3 mm) sheaths used for delivery of the 23- and 26-mm valves, respectively, require a minimal diameter of 7 mm for the 23-mm valve and 8 mm for 26-mm valve. The Sapien XT valve can be delivered through an 18-Fr sheath requiring a 6-mm iliac vessel. In this model, the valve is initially crimped on the shaft and then placed on the balloon after it is in the aorta (Figure 3).

The analyses of vessel tortuosity and calcification are most important to predict safe access through the iliofemoral vessels. Tortuosity without calcification that is straightened by a wire does not preclude the procedure. Tortuosity with marked calcification, especially in the bifurcation of internal and external iliac arteries, is of special concern since this area has no yield for expansion or movement: the internal iliac anchors the bifurcation down into the pelvis. Calcification is not well appreciated by angiography and requires CT for visualization. We have encountered many cases in which the iliofemoral angiogram showed adequate vessels for percutaneuos access, while the CT revealed severe calcification, thereby showing that the patient should be rejected for transfemoral access. Another method for measuring vessel diameter is IVUS, especially when there is discrepancy in the results of angiography and CT. IVUS does not allow analysis of calcification and has limited contribution in the final decision-making process. In general, we see larger diameters with IVUS than with CT.

Aortic Arch and Aortic Annulus and Coronary Ostium

Patients with extensive atherosclerosis of the aorta or large mobile protruding aortic atheromas are at higher risk for neurological events during the procedure. Atherosclerotic material can also be displaced from the aortic valve itself, can be carried up the aorta from the peripheral vessels, and embolized to the brain, thus resulting in ischemic stroke.¹⁰ Careful advancement of the retroflex delivery system around the arch, and the new RetroFlex 3 (Edwards Lifesciences) (Figure 4) can help prevent the mobilization of aortic plaques.

New devices are in development that will protect the arch vessels from particulate emboli. In cases of severe peripheral disease or a large atheroma in aorta, it is better to use transapical placement. Heavy calcification of the aortic valve facilitates positioning of the percutaneous valve, but has more risk of embolization and may cause problems for full and symmetric expansion of the valve. The aortic annulus diameter determines the valve size: 23- and 26-mm valves are now available for annular sizes of 18–21 and 22–25 mm, respectively. Precise annulus measurement before the procedure is crucial, especially in patients with borderline vascular access. Measuring the annulus by transthoracic echocardiography usually underestimates the correct diameter by 1–2 mm. TEE is now the standard for final determination of annular dimensions (Figure 5). In cases that may require a 26-mm valve with borderline iliac size, one must perform TEE before the percutaneous valve procedure to determine whether the transfemoral approach or transapical placement should be used. CT measurements of the aortic annulus are routinely obtained, but there is no standard dimension determined for this technique at this point. The distance between the aortic annulus and the coronary ostia shows a large variability and is independent of a patient’s height. The mean distance between the ostium of the left coronary artery and the base of the sinus of Valsalva is reported to have a wide variation ranging from 7.1 to 22.7 mm (mean 14.4 ± 2.9 mm). In almost 50% of the cases, the distance between the left coronary ostium and the annulus was smaller than the left coronary leaflet length. Despite this, the occurrence of coronary ostia obstruction has been reported in only a handful of cases.¹¹ It is important to evaluate the length between the inferior aspect of the annulus and the inferior aspects of the lowest coronary ostium for subsequent prosthetic aortic valve implantation (Figure 6).

Heights of the Edwards Sapien valve are 14.5 mm and 16 mm for the 23- and 26-mm valves, respectively. The proximal two-thirds of the stent are uncovered to allow for coronary perfusion should the prosthesis stent cover the coronary ostia. In cases of low left-main coronary artery origin, it is recommended to perform an aortogram during balloon valvuloplasty to determine if the aortic leaflet could obstruct the left main coronary artery ostium. Obstruction of the coronary ostium has been reported.

Procedure

Achieving optimal results with this novel technique requires the collaboration of a multidisciplinary team of experts in echocardiography, intensive care, anesthesiology, vascular surgery, radiology, cardiothoracic surgery, and interventional cardiology. Anesthesia. Most centers in the United States use general anesthesia during TAVI, tailored to achieve extubation soon after completion of the procedure. Because general anesthesia has intrinsic risks. In populations with critical AS and significant comorbidities, we prefer to perform the procedure without intubation, instead using conscious sedation given by an anesthesiologist.¹² Conscious sedation is administered with ketamine and propofol or with dexemetetomidine. The room is set up for general anesthesia, however, so if intubation becomes necessary, it can be implemented in a matter of seconds.

Our experience has been very positive, without downsides, and, we believe, with less risk to the patient. Access. The preprocedural assessment of the access vessels allows for the choice of the best technique. The femoral artery can be accessed percutaneously or by surgical cutdown. Percutaneous access is an excellent choice in patients with large vessels and mild-or-no calcification. Surgical exposure of the common femoral artery is another option that provides an extra level of security, especially when preassessment showed vessels with challenging characteristics. When choosing the percutaneous approach, it is important to choose the optimal puncture site for delivery of the large sheath. Close guidance by prior CT and bony structures is very helpful. Iliac angiography from the contralateral side and image masking on the fluoroscopic monitor also allows for precise puncture at the optimal site. After wire introduction and introduction of a 6 Fr dilator, the artery is preclosed using a single Prostar XL device (Abbott Vascular Devices, Redwood City, California) (Figure 7) or 2 Proglide 6 Fr devices (Abbott Vascular). A stiff wire is advanced through a diagnostic catheter, the 16 Fr to 15–18 Fr dilators are carefully advanced, then the 22- or 24-Fr delivery sheath is introduced. Heparin is given only at this time. We place the 22- or 24-Fr delivery sheath from the start and proceed to cross the aortic valve, dilate the valve, and then deploy the prostheses. We find that this method expedites the procedure greatly, as opposed to starting with 6–8 Fr sheaths to cross the valve, then placing 12–14 Fr sheaths for valvuloplasty, then finally using the 22–24 Fr sheaths. Avoid “pushing hard” if there is resistance when advancing the dilators. It is important to stop if the patient experiences pain (in those not anesthetized) or when the entire vessel is “being rocked” on fluoroscopy as one tries to advance the sheath. Different lubricants have being tried to facilitate advancement of the dilators or sheaths without any clear advantage. If the dilators or the large sheath does not reach the common iliac, one could consider accessing the common iliac retroperitoneally. In this case, we do not use a conduit: the sheath is tunneled from the original inguinal access point and enters the common iliac tangentially, with minimal trauma to the vessel (Figure 8), thereby simplifying the procedure. Valve delivery. The optimal plane for valve deployment is determined by 3-dimensional (3D) CT prior to the procedure, and by multiple low-contrast aortograms (with 10 ml of 50% diluted contrast); all three sinuses need to be aligned in one plane to be able to choose the optimal position of the valve. After crossing the aortic valve, an Amplatz super-stiff ST-1 (1-cm tip) is shaped with the nose of a plastic syringe, making a round loop with the stiff portion of the wire. This prevents the balloon, the nose cone, or others from advancing to the apex, and creates ventricular perforation (Figure 9).

The diseased native valve is first dilated by balloon valvuloplasty under rapid right ventricular pacing, which facilitates subsequent entry of the new valve and allows flow while the new valve position is adjusted for optimal delivery location. The optimal projection perpendicular to the annulus has already been chosen in cases of severe aortic regurgitation after the balloon valvuloplasty and the need for rapid valve delivery. The valve is crimped on a balloon just before implantation with a specially designed mechanical crimper to achieve a symmetrical low profile and to ensure retention on the delivery system. The balloon with the crimped valve is advanced with a delivery catheter (RetroFlex 3), which integrates a nose cone at the distal end to facilitate advancement of the delivery system around the aortic arch and eliminates resistance when crossing the native aortic valve. We ask the anesthesiologist to compress the common carotid arteries (not the bifurcation) while the valve assembly crosses the arch in order to minimize the chance of cerebral embolization. After the new valve has reached the left ventricle, the RetroFlex catheter is withdrawn into the ascending aorta and the valve is pulled to its optimal position, with 50–60% of the assembly on the ventricular side of the aligned sinuses. Valve positioning is based on fluoroscopy and aortography (Figure 10). TEE has limited ability to discriminate the valve from the balloon or the nose cone, but occasionally it can help with positioning, especially with coaxial valve aliment (Figure 11). In patients with marked hypertrophy of the septum or sigmoid septum, accurate positioning of the percutaneous valve may be hampered. During inflation of the balloon, the lack of space may squeeze the balloon towards the aorta before the stent/valve opens, and the valve will rise up with the risk of malpositioning or even embolization. To avoid this, we mount the valve lower in the balloon, which has resulted in less or no motion of the stent/valve during deployment (Figure 12).

The new RF 3 delivery device has minimal or no motion so this is no longer a major issue. After valve positioning, it is important to secure reliable pacing and to wait until the blood pressure falls below 50 mmHg prior to deployment and to continue pacing until the balloon is fully deflated. We use a 3.5 Fr screw-in pacer (Medtronic 6416-100 cm 3.5 F Pacer, Medtronic, Inc., Minneapolis, Minnesota) and find it more reliable than a 5 or 6 Fr regular pacing catheter. We perform prolonged inflation (5 sec after full inflation of the balloon) to obtain maximum expansion of the stent valve. This minimizes the presence of aortic regurgitation and minimizes the need for repeat inflations. We have demonstrated with quantitative coronary angiography (QCA) that the stent can grow as much as 2 mm with prolonged inflations; we also demonstrated recoil of 1 mm. After valve implantation, we perform aortography to assess aortic regurgitation. TEE has a central role in evaluating the degree of aortic regurgitation and its location; central aortic regurgitation is related to the stiff wire in the left ventricle. Paravalvular leak is quantified. If > 2+ and if the systemic diastolic pressure is low, repeat inflation of the balloon is required. Keys to minimizing aortic regurgitation include appropriate annulus sizing, good visualization of the annulus for valve deployment, full, prolonged inflation of the deployment balloon, and postdilatation of the valve. Patients with a bicuspid aortic valve are not optimal candidates for TAVI because the valvular orifice is elliptical and may preclude optimal seating of the stent valve.¹³

Valve migration during deployment is very rare, but has been reported. It is important not to remove the guidewire, as it prevents the valve from turning and occluding flow distally. A large balloon can be inserted through the valve and inflated in order to drag it past the left subclavian artery and deploy it. An aortic stent can then be deployed inside the valve to oppose the leaflets to aortic wall. Sheath removal. Iliac angiography is performed to ensure integrity of the iliofemoral vessels. In cases of surgical access, the access site is closed surgically. In percutaneous approaches, we inflate an 8–10 mm peripheral balloon at low pressure just below the iliac bifurcation while closing the access site with the Prostar or Proglide. The balloon is then released gradually. Final iliac angiography is performed to confirm vessel patency and runoff.

Clinical Results and Complications

The initial experience reported by Cribier et al14 included 27 patients who underwent TAVI with the Edwards valve: 23 with the antegrade approach and 4 with the retrograde approach. All were elderly with severe comorbidities. Procedural success was achieved in 75%. It was demonstrated that the mean valve area increased to 1.7cm2, with a small but significant improvement in global left ventricular function and with no obstruction of the coronary arteries. The 30-day mortality rate was 23% and the 30-day major adverse cardiovascular and cerebrovascular event rate was 26%. Patient survival was 63% by 6 months, and was determined by the severity of patients’ comorbidities. Later multicenter registries from the U.S. (REVIVAL II [tRans-catheter EndoVascular Implatation of VALves II] trial), European Union (REVIVE II [Registry of EndoVascular Implantation of Valves in Europe II]), and Canada (Canadian Special Access) included patients with valve areas 2 in 15 (3.2%), coronary obstruction in 3 (0.7%), and conversion to surgery in 8 (1.7%). At 30-day follow-up, death occurred in 29 (6.3%), stroke in 11 (2.4%), renal failure requiring dialysis in 23 (5%), permanent pacemaker in 31 (6.7%), vascular complications (access) in 14 (3%), and vascular complication (nonaccess) in 7 (1.5%).²¹ The long-term clinical results now available are based mostly on registry data. These registries have shown that there is a definitive learning curve. Webb et al¹⁵ reported that left ventricular function improves and mitral regurgitation diminishes significantly after valve implantation. Patients in whom mitral regurgitation improved had a lower left ventricular ejection fraction at baseline. The TAVI valve has superior hemodymnamic performance compared with the surgical bioprosthesis in terms of transprostetic gradients and preventsion of severe prosthesis patient mismatch, but was associated with a higher incidence of aortic regurgitation.²²

Mild perivalvular aortic regurgitation is reported in the majority of patients and is clinically inconsequential.¹⁷ In September 2007, the Edwards Sapien valve achieved the CE Mark in the European community, prompting diffusion of the technology and its use without the need of a formal research protocol, but with very strict inclusion and exclusion criteria. The largest ongoing randomized multicenter trial, entitled PARTNER (Placement of AoRTic traNscathetER valves), whose primary endpoint is 1-year mortality, just finished enrollment of > 1,200 patients mostly in North America and Canada. Two cohorts were included: inoperable patients randomized to a percutaneous valve or standard medical care, and high-risk operable patients randomized to a percutaneous valve (transfemoral or transapical) or surgical AVR. Trial results will answer many questions regarding the value, indications, and outcomes of the percutaneous aortic valve implants.

Summary

TAVI with the Edwards SAPIEN valve can be done safely and efficiently using a transfemoral retrograde approach. This technique offers potential treatment to patients with severe aortic stenosis who are currently undertreated or who undergo high risk surgery. Precise, detailed screening is the key for success in TAVI. The technique has rapidly evolved and is associated with a low number of complications (mostly vascular injury). A learning curve remains one of the most important predictors of procedural success and mortality. Today patients are selected based on their high surgical risk or inoperability. Significant decreases in sheath sizes may reduce complication rates and increase the availability of this technique for more patients. The results of the large, randomized, controlled trial may broaden the indications for TAVI.

From the Washington Hospital Center, Washington, D.C.

The authors report no conflicts of interest regarding the content herein.

Address for correspondence: Augusto D. Pichard, MD, Washington Hospital Center, 110 Irving Street, NW, Suite 4B-1, Washington, D.C. 20010. E-mail: guspichard@gmail.com