Author + information
- aDepartment of Cardiology, Swedish Heart and Vascular, Seattle, Washington
- bDepartment of Cardiology, CardioVascular Center Frankfurt, Frankfurt, Germany
- ↵∗Address for correspondence:
Dr. Sameer Gafoor, 550 17th Avenue, Suite 680, Seattle, Washington 98122.
Transcatheter aortic valve replacement (TAVR) for patients with pure native aortic valve regurgitation (NAVR) or failing regurgitant bioprosthetic surgical heart valves (SHVs) is a new frontier in the TAVR revolution. Each group of patients has distinct pathophysiological mechanisms to their disease process, and therefore medical management and procedural considerations are accordingly different. The strength of published data is considerably lower for NAVR or SHVs (nonrandomized registry or cohort data) than for patients with severe aortic stenosis (randomized controlled trials). The statistical limitations of such data, compounded by distinct differences between these 2 groups of patients, leaves many unanswered questions.
In this issue of JACC: Cardiovascular Interventions, Sawaya et al. (1) report on data collected from 18 centers participating in an international voluntary registry of patients treated with TAVR for pure severe NAVR or failing regurgitant bioprosthetic SHVs using Valve Academic Research Consortium-2 endpoints (2). Of note, the failing regurgitant bioprosthetic SHVs were only regurgitant in nature. Both older- and newer-generation heart valves were used, including the transapical JenaValve (JenaValve Technology, Munich, Germany). Seventy-eight patients with pure severe NAVR were included (average Society of Thoracic Surgeons [STS] Predicted Risk of Mortality of 6.7%), with predominant valves used being the Medtronic CoreValve (Medtronic, Dublin, Ireland) and the JenaValve. Device failure and need for second transcatheter valve occurred in 17% of NAVR patients, of which 50% were due to valve embolization. In addition, 14% of NAVR patients had post-procedural moderate-to-severe aortic regurgitation (AR). The all-cause mortality for NAVR patients was 14%, which was higher than that of the failing regurgitant SHV group. Of note, newer-generation valves did better than older-generation valves for the pure NAVR group.
A total of 68 patients (average STS of 7.7%) were included with failing regurgitant SHVs, with the majority of valves used to treat these patients being either the CoreValve or the SAPIEN XT (Edwards Lifesciences, Irvine, California). Device failure and need for second transcatheter valve happened in 10% of failing regurgitant SHV patients. Sixteen percent of the patients in the failing regurgitant SHV group had residual mean gradients above 20 mm Hg. The all-cause mortality in this group was 2%.
Certain important points come from this article. The percent of treated patients with pure AR is quite low (1.2% at enrolling centers), which may reflect lower prevalence or low confidence in procedural results. Lack of echocardiographic core lab and therefore self-reporting of leak, gradient, and success is a limitation. Despite the authors being from well-known, experienced centers, the success rate is much less than what we expect for aortic stenosis. The device failure occurs in roughly 10% to 20% of patients, of which half are due to valve embolization—this may be due to reverse hemodynamics or lack of calcium for fixation. Although the patients with failing regurgitant SHVs have had a prior sternotomy, the 30-day all-cause mortality was lower compared to NAVR patients, which points to a different risk profile. Device iteration has helped not only with aortic stenosis, but also with severe pure native AR: “new”-generation transcatheter heart valves (THVs) did better than “old”-generation THVs for NAVR by having a higher device success rate (85% vs. 57%), less need for a second valve, and lower incidence of moderate-to-severe residual AR.
For failing regurgitant SHVs, the natural comparator is the pure regurgitant patient outcomes from the VIVID (Valve-in-Valve International Data) registry (3), which included 139 patients with pure regurgitant failing SHVs. VIVID registry data demonstrated that regurgitant failing SHV patients treated with TAVR had the lowest rates of death from any cause, in comparison with those with stenosis or mixed modes of SHV failure. This was at 4.3% at 1 month compared to 2% in this study. Furthermore, pure regurgitant failing SHV patients in the VIVID registry had a 9.4% rate of at least moderate AR, whereas Sawaya et al. (1) reported a rate of 6% moderate or severe AR in their TAVR for failing SHV patients (3). Sawaya et al. (1) also reported that the rate of valve-related dysfunction in the TAVR for the failing SHV group was 22%, primarily driven by post-procedural mean gradient ≥20 mm Hg (3). Performing TAVR for a failing SHV <23 mm was significantly more likely to result in valve-related dysfunction via the mechanism of post-procedural mean aortic valve gradient ≥20 mm Hg. By comparison, data from the VIVID registry highlighted that valve label sizes <21 mm were associated with a statistically significant increase in death from any cause. Smaller surgical valves have worse mean aortic valve gradients with valve-in-valve TAVR therapy than larger valves do. As seen in the VIVID registry, Sawaya et al. (1) reported no statistical difference in rates of death or stroke between the different valve platforms used in this registry. This is an important point—that “old”-generation versus “new”-generation THVs did not translate to a significant difference in device success, early safety, or efficacy in patients with failing SHVs.
Prior management and selection of these patients is sine qua non for good outcomes. All patients should be rigorously examined for endocarditis or in case of regurgitant surgical valves, whether the regurgitation is paravalvular or central. Patients with AR have enlarged ventricles, which may expose them to concomitant functional mitral and therefore tricuspid regurgitation. The valves, annuli, and aortas are different in these patients. Failing SHVs benefit from having standard sizes, but may or may not have adequate coronary access (either due to height or small sinus), especially in case of stentless SHVs or those with concomitant root replacement. Unlike stenotic aortic valves, there is also no balloon valvuloplasty option as a diagnostic option in regurgitant aortic valve disease.
Procedural techniques are also different for AR, whether native or in failed SHVs. For native valves, there is poor visibility on fluoroscopy, and a pigtail in the cusp will lose contrast due to regurgitation into the ventricle. The thinner ventricles may expose patients to higher risk of perforation due to wire manipulation. Native valves may be bicuspid with irregular or absent calcium distribution and density, which is an issue for fixation. Ascending aorta enlargement also hinders ability to use the ascending aorta as a fixation mechanism. These are some of the main reasons for increased embolization potential. The increased rate of embolization requires careful technique and equipment to retrieve or replace valves when necessary. Concomitant mitral and tricuspid valvular regurgitation requires careful coordination with anesthesia and also a team capable of quick and effective hemodynamic support when necessary. Patients developing new left bundle branch block have worse hemodynamics, and whether this is cause or an effect is unclear but may be both. Specific procedural considerations are listed in Table 1 for TAVR in both NAVR and failing SHVs.
There has not been as aggressive a movement to iterate valves for AR as there has been for aortic stenosis. AR has different anatomic features compared to aortic stenosis that affect device success. Unlike aortic stenosis, where the predominant force is upward from the ventricle into the aorta, the force in AR is primarily from the aorta into the ventricle. In native AR, calcium score, density, and distribution are overall much less compared to aortic stenosis. If present, calcium can help with seal but also can lead to gutter effect and paravalvular leak. Sawaya et al. have proven that device iteration does help with native AR. The newer-generation devices, such as the Evolut R (Medtronic), SAPIEN 3 (Edwards Lifesciences), Lotus (Boston Scientific, Natick, Massachusetts), DirectFlow (Direct Flow Medical, Santa Rosa, California), and JenaValve, have unique characteristics. These include seal (SAPIEN 3 and Lotus), increased radial force (Lotus), and repositionability (Lotus, Evolut R, DirectFlow, JenaValve). In addition, the JenaValve uses active fixation to the aortic cusps, which may be useful here; this is similar to the ACURATE neo (Symetis SA, Écublens, Switzerland) upper stabilizers, Engager (Medtronic), and the Helio transcatheter dock (Edwards Lifesciences).
In general, balloon-expandable technology can provide a significant seal but may require overexpansion of the valve; this can theoretically increase the risk of pacemaker and also of rupture. Self-expandable valves allow for continual expansion. There are anecdotal reports of deliberately placing a self-expandable valve low in the left ventricular outflow tract and a balloon-expandable valve within this self-expandable valve to help expand this valve further. Additional stability and fixation may be even more useful in AR. The recently available Evolut Pro (Medtronic) is similar to the existing CoreValve Evolut platform but also has a mechanism to decrease paravalvular leak; this will be an interesting choice in this field. Specific valves for pure native AR help make the difference. The keys for a successful prosthesis in native AR seem to include minimally invasive access, repositionability, active fixation, sealing mechanism, ease of use, and durability.
The article also provides an important word of caution. Failing regurgitant SHVs can be treated with safety and efficacy by centers that perform TAVR, as long as proper case and valve selection is undertaken. However, even the most experienced centers have worse results with NAVR compared to patients with severe aortic stenosis. Success requires careful patient selection, patient optimization, complication management, and ability to use hemodynamic support quickly. Patients and families must be aware of and consent to the increased risks with this procedure compared to severe aortic stenosis as well as the off-label nature of the work, in addition to having the proper expectations for recovery; this should be reserved for the prohibitive patient population. With further device iteration, testing, and rigorous study, AR, whether native or in failed surgical valves, can have an alternative transcatheter option and indication.
↵∗ Editorials published in JACC: Cardiovascular Interventions reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Interventions or the American College of Cardiology.
Dr. Gafoor has served as faculty and proctor for Medtronic and Boston Scientific; on the scientific advisory board for Keystone Heart; as a site principal investigator for the Low-Risk Trial (Medtronic); as a site subinvestigator for the REPRISE trial (Boston Scientific); and as a site investigator for the REFLECT trial (Keystone Heart). Dr. Sharma has served as a site subinvestigator for the Low Risk TAVR Trial (Medtronic) and the REFLECT trial (Keystone Heart).
- 2017 American College of Cardiology Foundation
- Sawaya F.J.,
- Deutsch M-A.,
- Seiffert M.,
- et al.