Transcatheter Valve Implantation for Failed Surgical Aortic and Mitral Bioprostheses: A Single-Center Experience

Abstract: Objectives. We share our center’s experience with the use of transcatheter valvular therapies in the setting of failed bioprostheses. Background. As medicine continues to advance, the lifespan of individuals continues to increase, and current surgical valvular therapies begin to degrade prior to a person’s end of life. It is important to evaluate the efficacy and durability of transcatheter valves within failed surgical bioprostheses. Methods. Baseline characteristics, periprocedural complications, and long-term outcomes were collected and assessed in patients who received transcatheter valves for failing surgical aortic valve bioprostheses and mitral valve and ring bioprostheses from March 2011 to July 2018. Results. From our cohort of 1048 patients, we identified 45 individuals (4.3%) who underwent transcatheter replacement of a failed bioprosthetic valve or ring. Mean age at presentation was 80.8 ± 10.7 years and 75.5 ± 9.3 years, mean STS score was 9.3 ± 5.1 and 13.3 ± 8.7, and mean time to failure was 12.0 ± 5.2 years and 7.3 ± 4.5 years for aortic and mitral positions, respectively. At 1 year, time to event analysis suggested a 16.4% mortality rate for aortic replacement and 12.8% mortality rate for mitral replacement. Conclusions. We demonstrate outcomes from one of the largest single-center United States based cohorts of transcatheter replacements of failed surgical bioprostheses. Our center has demonstrated that it is feasible to pursue the replacement of failed surgical bioprostheses in the aortic and mitral positions with transcatheter valves given appropriate patient selection. 

J INVASIVE CARDIOL 2020;32(5):186-193. Epub 2020 March 11.

Key words: TAVR, TMVR, valve-in-ring, valve-in-valve

Transcatheter aortic valve replacement (TAVR) is a highly effective treatment strategy for the management of severe, native, symptomatic aortic stenosis in the intermediate, high, and prohibitive surgical risk patients. As the transcatheter valve technology and clinician experience evolved, TAVR began to be utilized for treating failing surgical bioprostheses in aortic position; furthermore, transcatheter valves present an interesting alternative to repeat open heart surgery for failed surgical bioprostheses or annuloplasty rings in mitral position (ie, valve-in-valve/valve-in-ring (ViV/ViR) therapies). As repeat surgical valve therapy has a high morbidity, mortality, and risk of surgical-related complications,^1,2 an alternative method to treat these failing bioprosthetic valves is needed. Surgical aortic and mitral bioprostheses are numerous in type, and transcatheter ViV/ViR implantation in the aortic and mitral positions requires meticulous comprehension of surgical prosthesis brands and materials, sizing, prosthetic hemodynamics, and associated failure modes. Thus, we reviewed our entire cohort of transcatheter ViV/ViR patients and report our associated outcomes.

Methods

From March 2011 to July 2018, we screened all patients that underwent transcatheter valve implantation at our Structural Heart Intervention Center at University Hospitals-Cleveland Medical Center (UH-CMC) and identified 45 consecutive patients who underwent aortic ViV (aViV; n = 26 patients) or mitral ViV/ViR (mViV/ViR; n = 19 patients). Patients were followed for a minimum of 1 year post procedure to July 2019. A multidisciplinary heart team reviewed preprocedural transesophageal echocardiogram (TEE), clinical parameters, and risk assessment (Society of Thoracic Surgeons [STS] score, frailty index, etc) for all patients. Patients who were deemed high or prohibitive surgical risk with favorable anatomy were selected for percutaneous intervention of the failing bioprosthesis. All patients signed an informed consent and all relevant information was entered into a UH-CMC-sponsored electronic database (RedCap). The institutional review board at UH-CMC approved this study. 

Procedure. All patients underwent a planning TEE and gated, low-contrast computed tomographic angiography (CTA) scan to evaluate the access, vasculature, prosthesis sizing, and surrounding structures. All surgical reports were obtained prior to the procedure. Transcatheter valve sizing was based on the type surgical prosthesis, CTA measurements, expert consensus, and the Bapat V Valve-in-Valve smartphone application for aortic (https://www.ubqo.com/viv) and mitral valves (https://www.ubqo.com/vivmitral). The appropriate fluoroscopic C-arm angle for the surgical bioprosthesis alignment was calculated prior to the procedure (Figures 1A and 1B). 

Aortic ViV. The aortic prosthesis characteristics are detailed in Supplemental Table S1 (supplemental materials available at www.invasivecardiology.com). In the majority of cases, we elected to use the CoreValve/Evolut R self-expanding, transcatheter valve and delivery system (Medtronic); however, if deemed appropriate, we utilized the Sapien XT/Sapien 3 balloon-expandable valves, the associated Commander delivery system, and properly-sized E-Sheath (Edwards Lifesciences). A standard minimalist approach (MA) was used for all transfemoral cases; however, in cases of alternative vascular access, general anesthesia and a hybrid operating room were utilized. The failing surgical bioprosthesis was crossed with a 6 Fr AL1 or AR 1 catheter and a soft, straight-tipped 0.035˝ guidewire, which was exchanged for an extra-small, preshaped 0.035˝ Safari wire (Boston Scientific) positioned in the left ventricular apex. Preballoon valvuloplasty was not performed. The transcatheter valve was then fluoroscopically positioned, aiming for shallow implant depths (CoreValve/Evolut R = 0-5 mm; Sapien XT = 0-2 mm) relative to the surgical bioprosthesis based on the radiopaque markers of the surgical valve.³ In cases of CoreValve/Evolut R, a very slow unsheathing of the valve with continued fluoroscopic guidance, switching between the optimal surgical bioprosthetic co-planar C-arm angle and the non-parallax view of the transcatheter valve was performed. Ventricular pacing at rates of 100-120 beats/min was utilized at the discretion of the physician to help stabilize valve deployment. For Sapien XT deployment, a two-step inflation was performed using rapid ventricular pacing at 180-200 beats/min to reduce cardiac output and motion to facilitate stable implantation. In all cases, a transthoracic echocardiogram (TTE) was used to evaluate postprocedure valve function, position, and stability. If necessary, based on hemodynamics and imaging, postdilation of the newly implanted transcatheter valve was performed. In cases of transfemoral access, femoral artery hemostasis was performed by closing two preplaced orthogonal Proglide sutures (Abbott). With alternative access sites, primary surgical closure was performed at the operator’s discretion.

Mitral ViV/ViR. The mitral prosthesis characteristics are detailed in Supplemental Table S2. In all cases, a Sapien XT/Sapien 3 balloon-expandable transcatheter valve and Ascendra transapical delivery system (Edwards Lifesciences) were used. Given transapical access, all procedures were performed under general anesthesia in a hybrid operating room. The full technique for transapical access has been described previously.⁴ The left ventricular apex was then palpated under direct visualization with TEE guidance to determine the appropriate point of access. The apex was then prepared with 4-6 orthogonally placed pledgeted sutures. Next, we punctured the apex with a standard 18 gauge, 0.035˝ access needle and advanced a soft J-tipped 0.035˝ guidewire into the left ventricle and retrograde across the mitral bioprosthesis into the left upper pulmonary vein, which was then followed by the insertion of a standard 6 Fr, 12 cm sheath. A wire exchange was performed for a 0.035˝, 1 cm, soft-tip Amplatz Super Stiff wire (Cook Medical). Then, we advanced the Ascendra delivery sheath into the left ventricle up to 4 cm based on fluoroscopic guidance and thereafter performed meticulous de-airing of the system. 

Notably, the chosen Sapien XT/S3 bioprosthesis was crimped onto the balloon delivery system with the valve skirt toward the left ventricle. Preballoon mitral valvuloplasty was not performed. The new valve was fluoroscopically positioned based on the type of surgical prosthesis present, similar to the aortic valve procedure described above.⁴ In all cases, rapid ventricular pacing was performed at a rate of 180-200 beats/min to reduce cardiac motion and output. The Sapien XT was deployed with a two-step balloon inflation to allow for fine adjustments during the implantation. Once deployed, the delivery catheter was withdrawn from the left ventricle and TEE was used to evaluate postimplant positioning and prosthesis function. If no further intervention such as postdilation was needed, a short burst of rapid ventricular pacing at a rate of 180-200 beats/min was employed to reduce blood loss through the apex while it was closed. 

Post procedure. If general anesthesia was employed, extubation was performed either post procedure or in the intensive care unit within 6 hours of completion. Patients were typically observed in the intensive care unit for 12-24 hours. Since the MA was used in the majority of aViV cases, patients were typically discharged within 1-3 days post procedure; however, for the mViV/ViR procedure, patients were discharged 1-3 days after extubation, assuming no complications occurred. All patients were given dual-antiplatelet therapy (81 mg of aspirin and 75 mg of clopidogrel) for 3 months; if warfarin was indicated for any reason, then triple therapy was prescribed until the international normalized ratio was therapeutic (>2.0) and then clopidogrel was discontinued. All patients received a postprocedure TTE and follow-up was performed at 1 week, 30 days, 6 months, and 1 year post discharge, with annual follow-up thereafter.

Definitions and endpoints. All patients were followed post procedure, as described above, and quantitative and qualitative data were prospectively collected. Procedural success for aViV and mViV/ViR implantation was defined per the Global Valve-in-Valve registry.⁵ Valve Academic Research Consortium-2 definitions were used for all other major endpoints and entered into our RedCap database.⁶ The primary endpoint for the current report was 30-day, 6-month, and 1-year all-cause mortality. Secondary endpoints included procedural success, length of stay, stroke, new pacemaker, postimplantation gradient, renal dysfunction, paravalvular leak, major bleeding, and vascular complications. New York Heart Association (NYHA) classification was only recorded if performed by any member of the heart team within 30 days of their anticipated follow-ups at 30 days and 1 year. Echocardiographic results were recorded at these same intervals if the patient presented for TTE.

Statistical analysis. Categorical variables are presented as numbers and percentages and compared using Chi-square test. Continuous variables are presented as means ± standard deviations and compared with t-test or Mann-Whitney U-test, as appropriate. Survival analyses were performed using the Kaplan-Meier method and outcomes were compared using log-rank test. All tests were two sided and P<.05 was considered statistically significant. The Statistical Package for Social Sciences, version 25 (SPSS) was used for all analyses.

Results

Between March 2011 and July 2018, a total of 1043 patients underwent transcatheter valve replacement therapy, with ViV/ViR interventions in 45 cases (4.3%). In the ViV/ViR cohort, a total of 26 patients (58%) underwent aViV and 19 patients (42%) underwent mViV/ViR procedures. The mean follow-up period was 343 days for the aViV group and 387 days for the mViV/ViR group. No patients were lost to follow-up.

Aortic ViV. Baseline characteristics are described in Table 1. The mean age at presentation was 80.8 ± 10.7 years, and the majority (61.5%) were men. Mean STS score was 9.3 ± 5.1. Seventy-seven percent had NYHA III/IV symptoms and mean left ventricular ejection fraction was 47.1 ± 14.7%. The mean time to bioprosthetic failure was 12.0 ± 5.2 years. Surgical prosthesis types included Carpentier-Edwards in 12 patients (46.2%), Medtronic Freestyle in 8 patients (30.8%), St. Jude Biocore in 2 patients (7.7%), Edwards Perimount in 1 patient (3.8%), Medtronic Mosaic in 1 patient (3.8%), and Sorin Perceval in 1 patient (3.8%). One patient had a previously implanted 26 mm Sapien XT failure due to a perforated leaflet. Labeled valve sizes were 21 mm in 7 patients (26.9%), 23 mm in 8 patients (30.8%), 25 mm in 5 patients (19.2%), 26 mm in 1 patient (3.8%), 27 mm in 3 patients (11.5%), and 29 mm in 2 patients (7.7%). The predominant mode of failure was stenosis in 15 patients (57.7%), followed by regurgitation in 6 patients (23.1%). In 25 cases (96.2%), a self-expanding prosthesis was used to treat the failing bioprosthesis. In 1 case of 27 mm Medtronic Mosaic with mixed failure, a 26 mm Sapien XT valve was implanted due to previous mitral valve replacement. 

Twenty-four of the 26 cases were performed using the MA. In 1 case, a failed 21 mm Medtronic Freestyle valve was treated using a 23 mm self-expanding CoreValve prosthesis, which was implanted via direct aortic route due to severely calcific iliac and subclavian vessels. In another case, a failed 27 mm Medtronic Freestyle valve was treated using a 29 mm self-expanding Evolut R prosthesis that was implanted via the subclavian route due to severely calcified iliac vessels. In-hospital outcomes are presented in Table 2. Procedural success rate was 100%. Mean gradient was significantly reduced by 17.6 mm Hg on average (34.7 mm Hg vs 17.1 mm Hg; P<.001). No patient had major stroke, major vascular complication, intraprocedural death, or emergency open heart surgery. One patient (3.8%) required permanent pacemaker insertion and 5 patients (19.2%) demonstrated trace to mild paravalvular regurgitation within 24 hours of valve replacement. No patient demonstrated moderate to severe paravalvular regurgitation. There were no in-hospital deaths. The mean length of stay was 4.0 ± 2.5 days. Twenty patients (77.0%) were discharged to home and 5 patients (19.2%) were discharged to a skilled nursing facility. 

At 30 days, 6 months, and 1-year, all-cause mortality rates were 0.0%, 10.8%, and 16.4%, respectively. Rates of readmission for any cause were 4.2% at 30 days and 9.5% at 1 year. There were no heart-failure related readmissions within the first year. Pacemaker implantation rate was 3.8% at 30 days, with no increase up to 1 year. No patient had a stroke up to 1 year post procedure. 

At time of discharge, 69.2% of patients were NYHA class II and 30.8% were class III. At 30 days, 73.1% reported symptom improvement post aViV. At this same time, 15.4% were NYHA class I, 73.1% were class II, and 11.5% were class III. At 1 year, 40.0% were NYHA class I, 20.0% were class II, and 40.0% were class III.

At 30-day and 1-year follow-up, mean gradient was unchanged at 15.3 ± 8.8 mm Hg and 14.8 ± 5.7 mm Hg. Ejection fraction was unchanged from post implantation (51.8 ± 18.0%) to 30-day (48.0 ± 14.1%) and 1-year follow-up (51.0 ± 9.7%). One patient (3.8%) developed moderate paravalvular regurgitation at 30 days post TAVR, two patients (7.7%) had mild paravalvular regurgitation at 30 days post TAVR, and 3 patients (11.5%) had resolution of their paravalvular regurgitation at 30 days post TAVR. At 1 year, the patient with moderate paravalvular regurgitation was graded as mild, and 1 of the 2 patients with mild paravalvular regurgitation was graded as mild. 

Mitral ViV/ViR. The mean age at presentation was 75.5 ± 9.3 years, and the majority (52.6%) were men. The mean STS score was 13.3 ± 8.7. Atrial fibrillation was seen in 63.2% of the cohort. NYHA III/IV symptoms were present in 94.7% of the cohort. The mean left ventricular ejection fraction was 47.5 ± 14.7%. Mean time to failure was 7.3 ± 4.5 years. Types of surgical prosthesis included the Medtronic Duran ring in 8 patients (42.1%), Medtronic Mosaic in 6 patients (31.6%), and the Carpentier Edwards in 5 patients (26.3%). Labeled valve/ring sizes were 27 mm in 5 patients (26.3%), 29 mm in 13 patients (68.4%), and 33 mm in 1 patient (5.3%). The predominant mode of failure was regurgitation in 9 patients (47.4%), followed by stenosis in 6 patients (31.6%) and by mixed stenosis with regurgitation in 4 patients (21.1%). 

A transapical route was chosen for 15 cases (78.9%), while 4 cases were performed via direct visualization, which requires open surgical access from either the left atrial or left ventricular perspective with manual placement of the balloon-expandable prosthesis and inflation without guidewire support. Procedural success rate was 100%. Mean gradient decreased from 11.6 ± 5.8 mm Hg to 7.3 ± 4.5 mm Hg (P=.01). In those patients with predominantly regurgitation or mixed mode of failure, the degree of mitral regurgitation decreased from moderate to severe (3-4+) to none to trivial (0+) in 13 patients. The effective orifice area (EOA) increased from 2.13 cm² to 2.53 cm². No patient had major stroke, permanent pacemaker, major vascular complication, or emergency open heart surgery. One patient demonstrated mild paravalvular regurgitation post procedure. 

The mean length of stay was 9.6 ± 10.1 days. One outlier was excluded from length of stay analysis due to concomitant left ventricular assist device and mViV procedure (length of stay post procedure was 48 days). There was 1 in-hospital death (5.3%) due to spontaneous pulseless electrical activity 5 days post procedure. Ten patients (52.7%) were discharged to home and 8 patients (42.7%) were discharged to a skilled nursing facility. 

At 30 days, 6 months, and 1 year, all-cause mortality rates were 5.3%, 12.8%, and 12.8% (Figure 2). Forty-five percent of patients were readmitted at 6 months, with no further readmissions up to 1 year. Nineteen percent of patients had heart-failure related readmissions at 6 months without any further readmissions. Pacemaker implantation and stroke rates were both 0.0% at 1-year follow-up. 

At time of discharge, 5.6% of patients were NYHA class I, 27.7% were class II, 55.6% were class III, and 11.1% were class IV. At 30 days, 86.7% reported symptom improvement post mViV/ViR, with 18.8% in NYHA class I, 43.8% in class II, and 37.4% in class III. At 1-year follow-up, 36.3% were NYHA class I, 18.2% were class II, and 45.5% were class III.

At hospital discharge, there were 4 cases (21.1%) of transcatheter valvular regurgitation and 1 case (5.3%) of paravalvular regurgitation. Mean ejection fraction was 47.6% post implantation and remained stable at 1 year (mean ejection fraction, 47.0%; P>.99). At 1 month, one patient (6.7%) developed mild paravalvular leak. At 30 days, no patient developed a new paravalvular leak or further progression of an existing paravalvular leak. 

Discussion

Over 250,000 heart valve prostheses are implanted worldwide.⁷ Over the last decade, there has been an increase in usage of surgical bioprostheses in both the aortic and mitral positions.^7,8 With a growing proportion of the population likely to outlive surgical bioprosthetic durability, the need to avoid reoperation in this high-risk cohort has bred an evolution in the management of these complex valvular disorders.⁴ Transcatheter ViV/ViR procedures represent a novel approach to the current standard of care management. We report our single-center experience with aViV and mViV/ViR procedures and demonstrate the following: (1) aViV procedures using MA and mViV/ViR procedures via TA access can both be performed in a safe and effective manner; and (2) clinical and echocardiographic outcomes were favorable for both procedure types at follow-up.

Aortic ViV. Our aViV procedure is typically performed using the MA, via a transfemoral approach in a standard cardiac catheterization laboratory with physician-directed, nurse-administered anesthesia. As has been shown previously, the MA is less resource intensive, is economically beneficial, and requires meticulous procedural planning.⁹ Our heart team ensures that valve identification, CTA evaluation, and sizing are agreed upon between the various team members. While the ViV smartphone application is a useful tool for initial guidance regarding the true internal diameter of the surgical bioprosthesis and the recommended sizing of transcatheter bioprosthesis to be used, its utilization is always to be combined with comprehensive CTA assessment. The CTA assessment aids in helping simulate ideal implant sizing, implant depth, and possible obstruction post implantation (ie, mitral valve impingement). In the case of fluoroscopically visible prosthesis frame or valve markers, the entire procedure at our center is performed without any contrast imaging during the valve deployment, contributing to low rates of acute kidney injury/dialysis post procedure (7.7% in the current report). The key is to understand the fluoroscopic overlay of the aligned surgical bioprosthesis (coplanar angle) and the non-paralax view of the transcatheter valve to ensure proper deployment. All of our aViV procedures are performed using a self-expanding prosthesis without pre-balloon aortic valvuloplasty. Our team’s avoidance of preimplantation valvuloplasty is related to the friability of surgical leaflets as compared to native tissue, which may predispose to leaflet tearing and valve frame disruption, resulting in hemodynamic instability and debris embolization/stroke.^10,11 Although this strategy is sound, the authors acknowledge that this may lead to a higher incidence of postdilation in the long term, especially with a higher rate of self-expanding prosthesis use. In the current report, postdilation was performed in 19.2% of our population for moderate or greater paravalvular regurgitation post deployment. Current registries demonstrate a lower overall rate of 12% for postimplantation valvuloplasty; however, when reviewing the CoreValve cohort alone, the rate (17%) was similar to ours.⁵ Since we are using the MA, our valve gradients and paravalvular regurgitation are assessed via TTE (or if necessary, postimplantation invasive aortic-left ventricular gradient).

The hemodynamic outcomes of the transcatheter valve are largely based on the size of the surgical bioprosthesis, with larger surgical sizes having better hemodynamics due to the nondistensibility of bioprosthetic rings.¹² However, we favor implanting self-expanding bioprostheses with supra-annular leaflets in aViV for three primary reasons: (1) postimplantation valve gradients are reduced (with concomitant EOA increase) because the new valve functions above the most constrained portion of the surgical annulus;^5,12 (2) the current state of evidence suggests wider (0-5 mm) implantation depth to obtain optimal hemodynamic results compared with the narrow landing zone of balloon-expandable valves (0-2 mm);³ and (3) the recapturability feature allows for fine positioning to ensure the appropriate implantation depth. As seen in our aViV cohort, the mean change in valve gradient was approximately 17.6 mm Hg, with final mean gradient of 17.1 mm Hg (P<.001). This is consistent with published registries and experiences.^3-5,12-15 Furthermore, these results are in a cohort where there is a relatively large proportion of 21 mm and 23 mm surgical valves (n = 15; 57.7.%), which reiterates the lower postprocedural gradients obtainable with supra-annular valve implantation.¹² We believe that the need to choose the appropriate transcatheter prosthesis prior to implantation is paramount. Since surgery is not an option for these patients, any degree of patient-prosthesis mismatch is unacceptable and may be associated with poor improvement in left ventricular ejection fraction and worse long-term survival.^16-19 That being said, in larger surgical valve sizes, the use of balloon-expandable prosthesis provides comparable results with respect to postimplantation gradients and clinical outcomes.¹⁵ 

 A special mention must be made for aortic regurgitation as a mode of failure for surgical bioprosthetic valves. In the Global ViV registry, approximately one-third had regurgitation as an etiology, which is again similar to our cohort (23.1% for regurgitation only and 19.2% for mixed stenosis and regurgitation).⁵ A key metric to procedural success is to leave no residual regurgitation and to avoid adding a secondary valvular lesion, ie, aortic stenosis. Although some degree of stenosis is likely due to the physical space available for the simultaneous presence of the surgical and transcatheter bioprosthesis, we believe that use of a supra-annular valve is likely to abate problems associated with high postvalvular implantation gradients. However, we do acknowledge that the use of balloon-expandable valves also has a role in this disorder.⁵

The ViV Global registry demonstrates that the 30-day and 1-year all-cause mortality rates are approximately 7%-8% and 17%, respectively, in all-comers with aortic bioprosthetic failure regardless of type of transcatheter valve implanted.^5,14 Our Kaplan Meier mortality rates were 0.0% at 30 days and 16.4% at 1 year. Our report is comparable, especially given our similar STS score of 9.3 ± 5.1 (ViV Global registry STS score = 10).¹⁴ Furthermore, our in-hospital events (acute kidney injury, vascular complications, major bleeding, stroke, and pacemaker implantation) were quite low and attributable to the use of the MA, procedural planning, evolving transcatheter valve technology, and mature operator experience. 

Valvular hemodynamics and function were improved post procedure, and the result remained stable at last available follow-up (mean, 343 days) in the majority of patients. There were no documented cases of prosthesis failure. While this is encouraging in the acute follow-up period, the durability of transcatheter prostheses for aViV procedures remains in question because of the recent adoption of this procedure. The next iteration of the Global ViV registry will likely shed light on transcatheter prosthesis longevity. Furthermore, the frequency of paravalvular regurgitation in our cohort improved over time. This is most certainly related to the use of a nitinol-based prosthesis that continues to expand even after implantation is complete. 

Mitral ViV/ViR. Currently, the U.S. Food and Drug Administration has approved the use of Sapien XT/Sapien 3 as the only transcatheter valves for implantation in the mitral position via the transapical route. This is tremendous progress in a field where the perioperative morbidity and mortality of redo cardiac surgery is >15% in patients with previous mitral valve surgery over the age of 75 years.²⁰ It is important to note that at this time, it is not technically possible to implant the CoreValve/Evolut/Evolut R due to the scaffolding size of the valve and theoretically deep implantation into the left atria. Again, procedural planning is vital with notable considerations. First, the saddle shape of the mitral annulus may result in paravalvular leaks due to the circular nature of the transcatheter valve; therefore, it is imperative that imaging assessment be performed to ensure optimal matching of surgical bioprosthesis with the transcatheter valve. For example, in the case of mitral annuloplasty, using deformable and/or complete rings is key to ensure effective sealing and to minimize embolization.⁴ Second, because of the mitral location, left ventricular outflow tract (LVOT) obstruction is a possible result of implantation.^4,21 CTA assessment of LVOT dimension, proximity of the surgical valve to the LVOT, the aorto-mitral angle, and virtual depiction of valve implantation are particularly germane to the procedure to ensure a safe deployment.^21-24 Furthermore, with recent advances in structural interventions, the use of percutaneous, intentional anterior mitral valve leaflet laceration (LAMPOON) may be indicated and a part of the valve implantation strategy to avoid LVOT obstruction^.25

Since our approach is exclusively TA, the procedure is performed in a hybrid operating room with general anesthesia and TEE guidance; however, an off-label transvenous transseptal approach has been described.²⁶ Our preference for TA access is due to our very developed surgical and interventional experience, coaxial alignment of the apex and mitral valve annulus, the short distance between the access point and mitral valve, and ease of delivery and control of the transcatheter valve. We have previously reported a novel implantation technique, ie, direct visual placement of the transcatheter valve. In this previously reported case, the patient had placement of a Heartmate II LVAD and mViV done concomitantly via a TA approach to minimize surgical time on cardiopulmonary bypass and the need for further reoperation.²⁷ This technique has been described in various case reports in native mitral stenosis secondary to the inability to safely decalcify the mitral annulus; however, this is the first instance of this procedure applied to an mViV implantation.^28,29 The use of this technique has allowed three additional valves to be placed in those who were not candidates for TA or transseptal transcatheter mitral valve replacement as both patient had significant left atrial thrombi. 

The results of our mViV/ViR cohort demonstrated that a greater proportion of our population was treated for regurgitant mitral rings. While the mode of failure is consistent with published reports, ring degeneration is typically second to bioprosthetic valve failure.^30-37 This likely reflects prior regional surgical bias for ring repair and thus may comprise a larger proportion of our overall mitral valve bioprostheses. Our average STS risk score was 13.3 ± 8.7, which is similar to prior publications.⁴ Our all-cause 30-day mortality rate was 5.3% (n = 1) at follow-up, which is lower than the currently reported rate of 8.5%.⁴ All cases except the in-hospital death demonstrated no significant in-hospital or 30-day complications related to the procedure. Of note, we had no patients with evidence of LVOT obstruction, despite a reportedly higher incidence with ViR,⁴ which again is related to procedural planning and achieving the correct implantation position and depth during deployment.^32,38 

At interim follow-up of 30 days and 1 year, no patients developed prosthesis failure, as evidenced by stable gradients, and there were no cases of greater than moderate paravalvular regurgitation. Similar to aViV, no evidence of long-term durability exists as of yet; however, in a review of mViV/ViR procedures, a total of 93 patients had no evidence of late valve failure.⁴ 

Study limitations. This is a retrospective cohort of a single-center experience that is subject to the biases associated with such reports. Although no surgical cohort was used for control and despite our limited numbers, we were able to demonstrate clinical rates comparable with published large ViV/ViR registries and experiences. These procedures are extremely complex and require extensive procedural planning and seasoned interventional/surgical experience, which may limit the generalizability of the results. We used self-expanding prostheses (CoreValve/Evolut R) in the majority of aViV procedures and exclusively used balloon-expandable prostheses (Sapien XT/Sapien 3) in all mViV/ViR procedures. Therefore, the use of other available prostheses in these procedures cannot be extrapolated. Of note, our institution prefers TA access to transseptal access for mitral valve replacements, as we feel this offers more control over valve deployment in terms of depth of implantation and possible LVOT obstruction due to valve implantation. Our longest follow-up was 4.5 years and shows no issues with transcatheter valve degeneration; however, the durability of these devices for use in ViV/ViR procedures is unclear to date and requires further study.

Conclusion

We have shown in a large, quaternary referral center that both aViV and mViV/ViR are viable options in patients who have a high surgical risk, resulting in favorable clinical and hemodynamic outcomes. It is our opinion that the findings demonstrated herein are due to a unified heart team assessment and comprehensive preimplantation planning. Larger and longer-term studies are necessary to elucidate clinical and durability outcomes. 

^*Joint first authors. 

From the ¹Department of Internal Medicine, University of California, San Diego, San Diego, California; ²Interventional Cardiology, The Heart Specialists of St. Rita’s, St. Rita’s Medical Center, Mercy Health, Lima, Ohio; ³the Department of Internal Medicine, University Hospitals Cleveland Medical Center, Cleveland, Ohio; ⁴the Valve & Structural Heart Disease Intervention Center, Harrington Heart and Vascular Institute, University Hospitals Cleveland Medical Center, Cleveland, Ohio; ⁵the School of Medicine, Case Western Reserve University, Cleveland, Ohio; and ⁶the Department of Cardiothoracic Surgery, Mayo Clinic, Jacksonville, Florida.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Attizzani reports consultant income from Medtronic. Drs Costa, Bezerra, and Simon report consultant income for Edwards Lifesciences. The remaining authors report no conflicts of interest regarding the content herein.

Manuscript submitted August 17, 2019, provisional acceptance given August 26, 2019, final version accepted December 3, 2019.

Address for correspondence: Guilherme F. Attizzani, MD, Assistant Professor of Medicine, Department of Cardiovascular Medicine, University Hospitals Cleveland Medical Center, 11100 Euclid Avenue, Cleveland, OH 44106. Email: guilherme.attizzani@uhhospitals.org

1. Jamieson WR, Burr LH, Miyagishima RT, et al. Re-operation for bioprosthetic aortic structural failure — risk assessment. Eur J Cardiothorac Surg. 2003;24:873-878.
2. Leontyev S, Borger MA, Davierwala P, et al. Redo aortic valve surgery: early and late outcomes. Ann Thorac Surg. 2011;91:1120-1126.
3. Simonato M, Webb J, Kornowski R, et al. Transcatheter replacement of failed bioprosthetic valves: large multicenter assessment of the effect of implantation depth on hemodynamics after aortic valve-in-valve. Circ Cardiovasc Interv. 2016;9:e003651.
4. Paradis JM, Del Trigo M, Puri R, Rodes-Cabau J. Transcatheter valve-in-valve and valve-in-ring for treating aortic and mitral surgical prosthetic dysfunction. J Am Coll Cardiol. 2015;66:2019-2037.
5. Dvir D, Webb J, Brecker S, et al. Transcatheter aortic valve replacement for degenerative bioprosthetic surgical valves: results from the Global Valve-in-Valve Registry. Circulation. 2012;126:2335-2344.
6. Kappetein AP, Head SJ, Genereux P, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. J Am Coll Cardiol. 2012;60:1438-1454.
7. Schoen FJ, Levy RJ. Calcification of tissue heart valve substitutes: progress toward understanding and prevention. Ann Thorac Surg. 2005;79:1072-1080.
8. Barnett SD, Ad N. Surgery for aortic and mitral valve disease in the United States: a trend of change in surgical practice between 1998 and 2005. J Thorac Cardiovasc Surg. 2009;137:1422-1429.
9. Attizzani GF, Alkhalil A, Padaliya B, et al. Comparison of outcomes of transfemoral transcatheter aortic valve implantation using a minimally invasive versus conventional strategy. Am J Cardiol. 2015;116:1731-1736.
10. Kirwan C, Richardson G, Rothman MT. Is there a role for balloon valvuloplasty in patients with stenotic aortic bioprosthetic valves? Catheter Cardiovasc Interv. 2004;63:251-253.
11. McKay CR, Waller BF, Hong R, Rubin N, Reid CL, Rahimtoola SH. Problems encountered with catheter balloon valvuloplasty of bioprosthetic aortic valves. Am Heart J. 1988;115:463-465.
12. Webb JG, Dvir D. Transcatheter aortic valve replacement for bioprosthetic aortic valve failure: the valve-in-valve procedure. Circulation. 2013;127:2542-2550.
13. Alnasser S, Cheema AN, Simonato M, et al. Matched comparison of self-expanding transcatheter heart valves for the treatment of failed aortic surgical bioprosthesis: Insights from the Valve-in-Valve International Data Registry (VIVID). Circ Cardiovasc Interv. 2017;10:e004392.
14. Dvir D, Barbanti M, Tan J, Webb JG. Transcatheter aortic valve-in-valve implantation for patients with degenerative surgical bioprosthetic valves. Curr Probl Cardiol. 2014;39:7-27.
15. Webb JG, Mack MJ, White JM, et al. Transcatheter aortic valve implantation within degenerated aortic surgical bioprostheses: Partner 2 Valve-in-Valve Registry. J Am Coll Cardiol. 2017;69:2253-2262.
16. Pibarot P, Dumesnil JG. Prosthesis-patient mismatch: definition, clinical impact, and prevention. Heart. 2006;92:1022-1029.
17. Head SJ, Mokhles MM, Osnabrugge RL, et al. The impact of prosthesis-patient mismatch on long-term survival after aortic valve replacement: a systematic review and meta-analysis of 34 observational studies comprising 27 186 patients with 133 141 patient-years. Eur Heart J. 2012;33:1518-1529.
18. Tasca G, Mhagna Z, Perotti S, et al. Impact of prosthesis-patient mismatch on cardiac events and midterm mortality after aortic valve replacement in patients with pure aortic stenosis. Circulation. 2006;113:570-576.
19. Urso S, Sadaba R, Vives M, et al. Patient-prosthesis mismatch in elderly patients undergoing aortic valve replacement: impact on quality of life and survival. J Heart Valve Dis. 2009;18:248-255.
20. Balsam LB, Grossi EA, Greenhouse DG, et al. Reoperative valve surgery in the elderly: predictors of risk and long-term survival. Ann Thorac Surg. 2010;90:1195-1200.
21. Blanke P, Naoum C, Dvir D, et al. Predicting LVOT obstruction in transcatheter mitral valve implantation: concept of the neo-LVOT. JACC Cardiovasc Imaging. 2017;10:482-485.
22. Bapat V, Pirone F, Kapetanakis S, Rajani R, Niederer S. Factors influencing left ventricular outflow tract obstruction following a mitral valve-in-valve or valve-in-ring procedure, part 1. Catheter Cardiovasc Interv. 2015;86:747-760.
23. Blanke P, Dvir D, Cheung A, et al. Mitral annular evaluation with CT in the context of transcatheter mitral valve replacement. JACC Cardiovasc Imaging. 2015;8:612-615.
24. Blanke P, Naoum C, Webb J, et al. Multimodality imaging in the context of transcatheter mitral valve replacement: establishing consensus among modalities and disciplines. JACC Cardiovasc Imaging. 2015;8:1191-1208.
25. Babaliaros VC, Greenbaum AB, Khan JM, et al. Intentional percutaneous laceration of the anterior mitral leaflet to prevent outflow obstruction during transcatheter mitral valve replacement: first-in-human experience. JACC Cardiovasc Interv. 2017;10:798-809.
26. Eleid MF, Cabalka AK, Williams MR, et al. Percutaneous transvenous transseptal transcatheter valve implantation in failed bioprosthetic mitral valves, ring annuloplasty, and severe mitral annular calcification. JACC Cardiovasc Interv. 2016;9:1161-1174.
27. Orasanu G, Al-Kindi SG, Robinson MR, et al. First-in-human experience with transcatheter mitral valve-in-valve implantation during left ventricular assist device placement. Circ Heart Fail. 2016;9:e003458.
28. Murashita T, Suri RM, Daly RC. Sapien XT transcatheter mitral valve replacement under direct vision in the setting of significant mitral annular calcification. Ann Thorac Surg. 2016;101:1171-1174.
29. Astarci P, Glineur D, De Kerchove L, El Khoury G. Transcatheter valve used in a bailout technique during complicated open mitral valve surgery. Interact Cardiovasc Thorac Surg. 2013;17:745-747.
30. Kliger C, Angulo R, Maranan L, et al. Percutaneous complete repair of failed mitral valve prosthesis: simultaneous closure of mitral paravalvular leaks and transcatheter mitral valve implantation - single-centre experience. EuroIntervention. 2015;10:1336-1345.
31. Cheung A, Webb JG, Barbanti M, et al. 5-year experience with transcatheter transapical mitral valve-in-valve implantation for bioprosthetic valve dysfunction. J Am Coll Cardiol. 2013;61:1759-1766.
32. Wilbring M, Alexiou K, Tugtekin SM, et al. Pushing the limits — further evolutions of transcatheter valve procedures in the mitral position, including valve-in-valve, valve-in-ring, and valve-in-native-ring. J Thorac Cardiovasc Surg. 2014;147:210-219.
33. Cerillo AG, Chiaramonti F, Murzi M, et al. Transcatheter valve in valve implantation for failed mitral and tricuspid bioprosthesis. Catheter Cardiovasc Interv. 2011;78:987-995.
34. Seiffert M, Conradi L, Baldus S, et al. Transcatheter mitral valve-in-valve implantation in patients with degenerated bioprostheses. JACC Cardiovasc Interv. 2012;5:341-349.
35. Seiffert M, Franzen O, Conradi L, et al. Series of transcatheter valve-in-valve implantations in high-risk patients with degenerated bioprostheses in aortic and mitral position. Catheter Cardiovasc Interv. 2010;76:608-615.
36. Bouleti C, Fassa AA, Himbert D, et al. Transfemoral implantation of transcatheter heart valves after deterioration of mitral bioprosthesis or previous ring annuloplasty. JACC Cardiovasc Interv. 2015;8:83-91.
37. Schafer U, Bader R, Frerker C, et al. Balloon-expandable valves for degenerated mitral xenografts or failing surgical rings. EuroIntervention. 2014;10:260-268.
38. Cheung A, Al-Lawati A. Transcatheter mitral valve-in-valve implantation: current experience and review of literature. Curr Opin Cardiol. 2013;28:181-186.