Aortic regurgitation after transcatheter aortic valve implantation

Keywords: :

• aortic regurgitation
• CoreValve
• Edwards SAPIEN
• transcatheter aortic valve implantation

Aortic valve stenosis (AS) is the most common valvular disease, the incidence of which is increasing with an expanding aging population. Symptomatic severe AS is associated with a poor prognosis and a 50% yearly mortality in patients presenting with syncope. Treatment of severe AS has traditionally been with surgical aortic valve replacement (AVR), which has been demonstrated to be an effective treatment with associated improvements in symptoms and survival. However, >30% of patients with symptomatic severe AS are precluded from surgical AVR due to high surgical risk [1]. The introduction of transcatheter aortic valve implantation (TAVI) over recent years has transformed the treatment of severe AS in such patients, as this does not require thoracotomy and can be performed without general anesthesia. A number of studies have demonstrated that TAVI in this high-risk population is associated not only with good procedural outcomes but also improvements in clinical outcomes [2]. Despite this, a significant proportion of patients does not experience the full clinical benefit that can be associated with TAVI, something that has largely been attributed to post-procedural aortic regurgitation (AR). The reported prevalence of this varies, but on the basis of recent studies it appears to range from 6 to 21% for moderate or severe AR [3]. Here we discuss the mechanisms of AR following TAVI, its clinical impact as well as novel devices that aim to reduce this.

Mechanisms of AR following TAVI

The most important anatomical structure for successful TAVI is the aortic annulus. In contrast to surgical AVR where the surgeon has clear view of the outflow tract and thus can directly assess valve size, the choice of valve size in TAVI is based only on images generated by echocardiography, angiography and computed tomography (CT) images. Thus, inappropriate valve sizing can result in a too small prosthesis and therefore in paravalvular AR. In addition, valve placement in TAVI takes place whilst the heart is beating, and although rapid pacing helps to maintain valve position, displacement during implantation (either too high or too low) can still occur also resulting in AR. With increasing experience and improvements in cardiac imaging, however, the role of these factors in causing paravalvular AR after TAVI has been significantly reduced. Nevertheless, AR remains an important issue, with extensive valve calcification commonly being responsible for this [4]. Unlike in surgical AVR, where the native valve is excised and the prosthetic valve is sutured on a decalcified annulus, in TAVI the native valve is pushed away toward the annulus, and therefore, in the presence of extensive calcification, this can lead to asymmetrical valve expansion and paravalvular leaks. AR is also more likely to occur when the angle between the aorta and left ventricular outflow tract is large. Greater angle between the outflow tract and ascending aorta is a predictor of post-procedural AR in patients treated with CoreValve [5]. It can also be the result of balloon dilatation following implantation, which is occasionally performed in an attempt to improve apposition of the prosthetic valve to the aortic annulus, and therefore reduce paravalvular AR. Although this can have the desirable effect, it can, albeit rarely, damage the valve itself, leading thus to central AR. Central AR in the longer term can also result from valve leaflet malfunction and infective endocarditis.

Clinical impact of AR following TAVI

In AS, the increased obstruction between the left ventricle (LV) and the aorta leads to compensatory LV hypertrophy, which, although helps to maintain cardiac output at its early stages, later leads to myocardial fibrosis, impaired relaxation and diastolic dysfunction. In such cases, the LV is less well equipped to cope with the regurgitant volume of AR, even if this is not severe, leading to increases in LV end-diastolic pressure, pulmonary edema and elevated pulmonary arterial pressures in the longer term. It is therefore not surprising to see recent evidence suggesting that the presence of significant AR after TAVI is associated with worse clinical outcomes. In the Italian multicenter study of 663 patients undergoing TAVI with the CoreValve ReValving system (Medtronic, Minneapolis, MN, USA), moderate or severe post-procedural AR, as assessed by echocardiography, occurred in 21% of the patients and was found to be an independent predictor of mortality at 1 year [6]. Similar results were reported from an 870-patient cohort from the UK Transcatheter Aortic Valve Implantation Registry, in which 52% of patients were treated with the CoreValve prosthesis and 48% with the Edwards SAPIEN valve (Edwards Lifescience, Irvine, CA). Aortic regurgitation diagnosis in this study was based on post-procedural aortography, which also demonstrated that severe post-procedural AR was more common with the CoreValve (17.3%) rather than the Edwards SAPIEN (9.6%) prosthesis (p < 0.001) [7]. The large study by Gilard et al. (n = 3195) also demonstrated that significant AR (AR grade ≥ 2) following TAVI was an independent predictor of mortality at 1 year [8]. The large meta-analysis by Athappan et al. consisting of 12,926 patients confirmed that the incidence of significant regurgitation (moderate/severe) after TAVI was relatively high at 11.7% with moderate/severe AR associated with early and late mortality and mild AR possibly associated with the late mortality [4]. These results are in agreement with the recently reported 3-year results of the PARTNER ‘Placement of AoRTic TraNscathetER Valve’ trial that demonstrated that not only moderate or severe AR but also mild paravalvular and total post-procedural AR is associated with increased subsequent mortality [9]. The adverse effect of AR on long-term survival was also evident in the study by Toggweiler et al. from an 111-patient cohort who had undergone TAVI with either the Cribier-Edwards or Edwards SAPIEN valve, who demonstrated that 5-year survival is reduced when TAVI is associated with at least moderate paravalvular leak [10]. The fact that even mild AR impacts on survival should not come as a surprise, as unlike healthy hearts, patients with severe AS already suffer from stiff ventricles and are thus less likely to able to accommodate even slight increases in volume load. It is likely that patients with either no preexisting AR prior to TAVI (‘AR-naïve’) or significantly larger grades of AR post-procedure (e.g., from mild to severe) are more vulnerable to the adverse effects of volume overload as compared to patients with preexisting AR and no significant changes in grade after TAVI, as the LV in the latter case is already ‘accustomed’ to volume overload. In patients therefore with severe AS who are ‘AR-naïve’, the occurrence of AR, even if this is mild, can exert effects similar to acute severe AR in a normal patient.

Prevention & treatment of AR following TAVI

The first step in preventing paravalvular leaks following TAVI is an accurate assessment of annulus size and thus valve size selection. As different imaging modalities have been demonstrated to provide similar but not necessarily the same results [11], evaluation of aortic annulus size using both CT and echocardiography, either transoesophageal and/or 3D, can help to prevent errors in valve sizing. Aortography during balloon valvuloplasty should also be used in cases where uncertainties exist about valve size, since if significant AR is present during this maneuver, valve upsizing may be required. In patients with a large angle between the left ventricular outflow tract and ascending aorta, the Edwards SAPIEN valve may be a better choice as compared to the CoreValve. The treatment choice for AR following implantationlargely depends on the underlying cause. In cases of valve underexpansion or poor valve apposition to the annulus, balloon post-dilatation should be used [12]. In cases where the valve is positioned too high or too low, a second valve can be implanted, and in the case of the CoreValve system, a snare can be considered to pull the valve up when this is too low.

The best treatment, however, for paravalvular AR is to avoid this altogether, and although careful planning and correct valve sizing are important, this sometimes is not possible with the early-generation transcatheter valves. Newer generation valves aim to address this limitation by incorporating features in their designs that specifically tackle paravalvular leaks. The Direct Flow Medical Aortic Valve (Direct Flow Medical, Santa Rosa, CA) consists of a bovine pericardial tissue valve mounted between two inflatable polyester rings that is delivered through an 18Fr system via the transfemoral approach. The two rings adapt around the native aortic annulus and outflow tract, creating a seal on either side of the valve, minimizing thus paravalvular AR. Furthermore, the two rings can be deflated and reinflated prior to final deployment, achieved through the exchange of contrast medium in the rings with the hardening resin, allowing small adjustments in valve positioning that may help to reduce any residual AR. The results of the DISCOVER (Direct Flow Medical^® Transcatheter Aortic Valve System for the Treatment of Patients with Severe Aortic Stenosis) CE Mark trial, which recruited 100 patients and were presented in this year’s EuroPCR, demonstrated that 99% of patients experienced mild or less AR and 73% of the patients zero or trivial AR [13]. The Sadra Medical Lotus™ Valve (Boston Scientific, Natick, MA) is another new generation TAVI valve delivered through the transfemoral route with features to minimize AR. This valve is made of bovine pericardium mounted on a braided nitinol stent structure expanding on the aortic annulus as it shortens. At the lower part of the prosthesis, an ‘adaptive seal’ is present making the valve more conformable to annular anatomy. The 18Fr delivery system allows the valve to be recaptured and repositioned if necessary. The first-in-man (FIM) feasibility trial REPRISE I (Repositionable Percutaneous Replacement of Stenotic Aortic Valve through Implantation of Lotus™ Valve SystEm) enrolled 11 patients, with 9 of them no AR and 2 only mild AR at 90-day follow-up [14]. The REPRISE II CE Mark study currently underway will recruit 120 patients worldwide. Results from the first 60 patients demonstrated that at 30-day, 79% of patients had trace or no AR, with only 1 patient experiencing moderate AR [15]. The Portico valve (St. Jude Medical, St. Paul, MN) has a tissue-cuff in the lower part of the prosthesis to reduce paravalvular AR. The large cells of the valve frame at the annulus position, together with the presence of more tissue and less metal help to reduce paravalvular leaks by making the valve more conformable and less likely for the stent struts to rest against calcific nodules. The valve is also resheathable, allowing therefore repositioning and retrieval. This valve can be delivered both via the transfemoral and transapical routes through an 18Fr and 24Fr system, respectively. The 23 mm valve is available for clinical use in Europe. In the 10 patients included in the FIM study, moderate AR was present in 1 patient [16]. Further trials are currently underway with the 25 mm valve. The newer generation balloon expandable Edwards valve, Edwards SAPIEN 3 (Edwards LifeSciences, Irvine, CA) although not retrievable or repositionable can minimize AR, thanks to an external sealing ring. This valve can be delivered via a 14Fr sheath. The self-expanding Edwards CENTERA, unlike the SAPIEN 3, is repositionable and is delivered through a motorized system allowing single operator use. In the FIM study of 15 patients, only trivial AR was noted at 1-year follow-up [17]. In addition to these valves all of which can be implanted via the transfemoral route, other new generation valves delivered through the transapical route have incorporated features to reduce the incidence of paravalvular leak. The Medtronic Engager™ (Medtronic, Minneapolis, MN) and JenaValve (JenaValve, Munich, Germany) valves are mounted on a nitinol self-expanding stent with skirts in their lower parts (polyester and pericardial, respectively) aimed to address this need. In the case of the JenaValve, the native valve leaflets are clipped between the feelers and the base of the prosthesis reducing the risk of paravalvular leak. The 1-year results of the CE Mark study, which included 67 patients, were encouraging with 67.6% of patients experiencing no AR and only 8.8% grade 2. There were no cases of > grade 2 AR [18]. Another transapical valve with promising results is the Symetis Acurate TA™ valve (Symetis, Lausanne, Switzerland), which possess a Dacron skirt at the lower valve crown providing a seal at the level of the annulus. In the SAVI (Symetis Acurate TA™ Valve Implantation) registry of 250 patients, only 2.7% of patients demonstrated paravalvular AR ≥2 at 30-day follow-up [19]. A transfemoral system for this valve is currently under clinical evaluation.

Conclusion

TAVI has revolutionized the treatment of severe symptomatic AS in high-risk surgical patients. As experience developed, the risk of complications such as vascular access bleeding, annular rupture and the occurrence of conduction abnormalities decreased. This editorial has pointed out the presence of residual AR as one of the last surviving limitations, which nevertheless impacts on short- and long-term outcomes. The etiology of AR in these cases is largely due to the presence of calcifications that prevent complete apposition of the implanted valve to the annulus. New generation TAVI valves have incorporated specific features to their design, aiming to address the issue of paravalvular AR.

Future prospects

If the expectations linked to new generation valves such as repositioning and absence of residual AR prove to be consistent, we can expect TAVI to become more safe, with further improvements in clinical outcomes and applicable to intermediate- and low-risk patients. We also expect that in selected patients, TAVI will have a role in treating native AR. We envisage that this in conjunction with an aging population and a large ‘TAVI-suitable’ population number that remains to be treated will result in a significant increase in the number of TAVI procedures performed each year [20]. We also expect that in selected patients TAVI will have a role in treating native AR.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.