One-Year Outcome after Transcatheter Aortic Valve Replacement for Aortic Regurgitation: A Single-Center Study

Background Presently, there are limited reports in the literature on the post-operative (mid-term) clinical outcome for pure Aortic Regurgitation (AR) following Transcatheter Aortic Valve Replacement (TAVR). Methods Between and 2019, a total of 134 high-risk patients with pure, symptomatic severe AR patients were enrolled in the current study. The outcome was assessed according to the VARC-2 criteria. Procedural results, clinical outcomes, and the patients' hemodynamics for a period of 1-year were analyzed. to a lack of valve calcication and the simplicity and reliability of the THV implantation in patients.


Introduction
Transcatheter aortic valve replacement (TAVR) has rapidly become a therapy for severe aortic stenosis (AS) in elderly patients 1,2 . There is a growing number of elderly patients with a high-risk pro le suffering from aortic valve regurgitation (AR) and an abysmal poor prognosis with medical treatment 3 . However, there are still some challenges in treating AR with TAVR due to the increased risk of valve embolization, migration and paravalvular leakage (PVL) in the absence of aortic annular calci cation. Despite surgical aortic valve replacement (SAVR) being the gold standard for treating AR 1 , with an annual mortality rate of 10-20% 4, 5 , a few elderly patients refuse surgical operations due to the involved surgical risk 3 .
The initial application of TAVR for AR using rst-generation transcatheter heart valve (THV) showed high rates of procedural complications and low device success rates 6,7 . With the evolution of THV design, technical improvements were made to overcome the procedural challenges in treating AR. The J-Valve is a self-expanding, porcine aortic valve stitched within a nitinol frame. Novel features of the THV are its 2piece structure, longitudinal anchoring mechanism, and automatic anatomical ability to reposition 8 .
Previous research has demonstrated low mortality and stroke and excellent hemodynamics 9,10 .
In 2017, the clinical application of J-Valve was approved by the China Food and Drug Administration for AS and AR patients. TAVR treatment for AR has not yet been validated for a large-scale clinical application. Herein, we report our 1-year institutional experience with transapical TAVR using the J-Valve for the treatment of pure AR patients.

Study design and patient characteristics
The present case-series is a single-center, retrospective, observational study.  11 . Patients with mixed valve disease or a peak aortic jet velocity of ≥ 2.5 m/s were excluded from the study 12 . The inclusion criteria for patients in the study were: (i) patients' age ≥60 with NYHA functional class of -; and (ii) patients were assessed to have severe AR with an indication for TAVR by a multidisciplinary heart team. All patients signed written informed consent for prospective data acquisition and follow-up examinations.

Endpoints
The primary endpoint of the study was overall mortality at one and twelve months. Cardiac mortality and procedural results were also considered at the same time points. Device success and the clinical safety endpoint were evaluated and determined according to the Valve Academic Research Consortium-2 (VARC-2) criteria 13 . Other endpoints included device-related complications, and echocardiographic assessment of the valve and cardiac function at post-procedure. The severity of the PVL was qualitatively assessed and graded using TTE 14 .

Study device
The J-Valve is a self-expandable, porcine aortic valve stitched within a nitinol frame with 3 U-shape anatomically orientating claspers ( Figure 1A). All procedures were performed with the patient under general anesthesia delivered via a transapical route. Details of the implantation procedure were previously reported 9 . The stent implantation was performed in two-stages during the procedure.
I. The graspers can be fully released and wholly seated in each sinus of the tricuspid aortic valve before valve deployment ( Figure 1B), enabling correct valve xation in a supra annular position. Precise positioning of the clasper can also be applied to the bicuspid aortic valve (BAV) ( Figure 1C).
II. The valve's prosthesis is then retrieved into the aortic sinus and released without rapid ventricular pacing ( Figure 1D). Five valve sizes were available during the study (21-, 23-, 25 -, 27-and 29-mm). The diameter was measured as the largest possible diameter during systole using the Multidetector computed tomography (MDCT) perimeter 15 .
Multidetector computed tomography (MDCT) perimeter-based diameters determined the choice of valve size (21 mm, 18 to 20.9 mm; 23 mm, 21 to 22.9 mm; 25 mm, 23 to 24.9 mm; 27 mm, 25 to 26.9 mm; and 29 mm, 27 to 29 mm). MDCT was performed on all of the patients both before and after the procedure ( Figure 2).

Procedures
Aortic annulus, root, and valve morphology were assessed using both contrast-enhanced MDCT and transesophageal echocardiography (TEE). The THV severed to anatomically orient the position for optimal implantation along with the help of the clasper (Figure 3). Evaluations of valve function and quanti cation of residual aortic regurgitation after TAVR were performed by TEE and angiography ( Figure  3, 4).
The procedure was performed in a hybrid operation room under general anesthesia while a full cardiopulmonary bypass circuit was on standby. A pigtail catheter was advanced into the ascending aorta via the right femoral artery and an aortogram was performed. TEE was used for the evaluation of the valve pathology. A temporary pacemaker was placed, and a 4 cm incision in the coastal space at the heart's apex was made. Aortic root angiography was used to identify the aortic sinus and annulus. A J-Valve was crimped into the Ausper-AS delivery system. The delivery system was inserted into the left ventricle through the apex and advanced into a supra-annular position under uoroscopic guidance.
The three U-shaped anchor rings were ultimately deployed, pulled down and tactile feedback was checked to ensure that the three anchors were inside the aortic sinus ( Figure 3A/E). Then aortic root angiography was reperformed. Next, the top part of the delivery system in which the valve was stored was retrieved back gently into the annular plan ( Figure 3B/F) and deployed without rapid ventricular pacing ( Figure 3 C/G). The aortic root angiography was used to monitor PVL ( Figure 3D/H). The functionality was also con rmed via TEE.

Statistical analysis.
Continuous variables following normal distribution were presented as mean ± standard deviations. Nonnormally distributed variables were presented as median and range. Categorical variables were presented as counts and percentages. All data were analyzed with SPSS 23.0 statistical software (SPSS23, Chicago, IL, USA). A p-value of 0.05 was considered statistically signi cant. A 1-year Kaplan-Meier estimate of all-cause mortality was calculated for patients.

Results
Baseline characteristics A total number of 134 patients with severe AR were enrolled. The Mean age was 73.1±6.4 years and 25.4% were female ( Table 1). The mean risk score according to the Society of Thoracic Surgeons (STS) was 9.8±5.3%, while 131 patients (97.8%) were in New York Heart Association (NYHA) functional class III or IV. Four patients had previous cardiac surgery. Five patients had undergone a previous permanent pacemaker. In the present study, morbidity rates were higher for chronic lung disease (n=73, 54.5%), peripheral vascular disease (n=64, 47.8%), and cerebrovascular diseases (n=67, 50.0%).
Procedural characteristics and prosthesis sizing.
The procedure was successful in 97.1% (130/134) and device success was 96.3% (129/134). Four patients were converted to open-heart procedures during the operation (one coronary obstruction, two valvular embolisms, and one valve migration) ( Table 2). We updated our protocol in subsequent cases to prevent this complication, including a mandatory repeat of root angiography from con rming the clasper's position and con guration after deployment. There was no valve dislodgements after that update. In one patient that mismatch led to moderate PVL and cardiac insu ciency six days after the operation.
The mean aortic annular diameter was 24.9±2.7 mm, resulting in a median prosthesis oversizing by area of 6.4%. The 21-mm, 23-mm, 25-mm, 27-mm, and the 29-mm prosthesis valves were implanted in 2.2%, 10.4%, 23.1%, 41.0%, and 23.1%, respectively in the enrolled patients. Deformation dynamics of the aortic valve annulus in different valve pathologies may vary in different forms of AS. Caution should be used during annular sizing in patients undergoing TAVR. The new permanent pacemaker implantation (PPMI) rate was 7.4%. Myocardial infarction, cerebrovascular events, coronary obstruction requiring intervention, life-threatening bleeding and endocarditis were not observed. The median time of stay in the intensive care unit was one (1-1) day. The average postoperative hospital stay was six (5-8) days.
Echocardiographic assessments of the heart function at baseline and follow-up are shown in Table 3. The mean gradient pressure remained stable at one month, six months and twelve months (8.2±3.  Table 4. No patient was missing in the present study. The 30-day all-cause mortality rate was 2.9%, and no major stroke occurred in any patient. In these patients, 1-year all-cause mortality was 7.4% ( Figure 5). Landmark analysis demonstrated that the most mortality occurred after 30 days of the procedure (Figure 6). Three patients had a stroke, resulting in a 1-year stroke rate of 2.2%, where two patients died. One patient had a type-A aortic dissection at six months follow-up, while one patient underwent reoperation due to prosthetic valve thrombosis at seven months postoperatively, although the latter refused further surgical treatment. New PPMI was required in twelve patients (8.9%).

Discussion
We report 1-year outcomes from a single-center experience evaluating the J-Valvefor TA-TAVR in patients with AR. The key ndings include 7.4% and 2.2% 1-year mortality and stroke rate, respectively. Echocardiographic measurements con rmed adequate hemodynamic function with a signi cant improvement in LVEF, a reduction in LVEDD, LVESD, a low rate of residual leakage after TAVR, and excellent aortic valve hemodynamics, with low mean gradients of 8.2±3.2 mmHg maintained to 1-year follow-up.

Procedural Outcomes
Severa observational studies show that TAVR is a viable option for patients with pure AR 6,7,16,17 . It is well known that the rst-generation THV primarily relying on calci cation of the native lea ets for su cient anchoring of the expanded prosthesis. Patients with AR have more complex and variable anatomy, and the lack of calci cation may lead to inaccurate positioning and di culty of anchoring, or even worse, valve embolization or residual PVL 18 . TAVR using the new-generation devices was associated with improved procedural outcomes in treating patients with AR 19,20 . Importantly, with the second-generation valves and advanced technique J-Valve has even been certi ed in the indication AR. With a novel xation mechanism associated with a signi cantly higher procedural success rate, these devices are a reasonable option in AR patients 21,22 . In the present study, the device's success is consistent with previous reports of the same valve in patients with AS 23 . The device's success was 96.3%, which corresponds to the J-Valve's initially reported experience for AR treatment, which was 97.6% 9 . A recent meta-analysis has reported that, based on nineteen studies, with a total number of 988 patients, the rate of devices success was 86.2% (78.8%-92.2%) 24 . Our study demonstrated a higher success rate and a lower mortality rate than the studies mentioned above, indicating a novel device's viable safety pro le. However, there was no difference in the device success rate between both devices [J-valve and JeneValve (96.3% vs. 96.8%)] 16 .
The feelers in JenaValve have a rigid connection with the support frames and J-valve with mobile nitinol graspers, which are both designed to be placed into the sinus of the aortic root to achieve an anatomically correct position. The unique design make it possible to ensure the optimal positioning of the valve stent after deployment.

Mortality and Stroke
In the present study, the low 1-year mortality rate of 7.4% was consistent with the 1-year rate of 4.7% in forty-three high-risk patients from the China Trial 9 and a 5.5% mortality of TA-TAVR procedures in Germany in 2014 25 . The 1-year mortality rates reported with the J-Valve device are lower than the rstgeneration reported in other studies, namely the CoreValve, 31% in 26 high-risk patients 7 and 21.4% in the other CoreValve study (n=43) 6 . Silaschi et al reported transapical TAVR with higher a 1-year mortality rate (20.1%) than that the present study 26 . The initial German experience 6-month mortality was 19.3% with JeneValve for the treatment of AR 16 . Compare to AS, although AR is younger and lower mean STS score, howerver the mortality was no differences 27 . These differences suggest differing pathophysiology of the larger left ventricle and the lower LVEF after treatment of AR is quantitatively and qualitatively different from AS 28 . Notebly, a recent meta-analysis of AR patients who underwent TAVR showed that the one-year mortality is 25% 24 .
The incidence of disabling stroke at 1-year (2.2%) in the present study was consistent with the 2.3% reported in the China clinical Trial 9 . At 1-year, stroke was 4.7% in CoreValve experience and 3.3% 7 in JeneValve of the JUPITER registry 26 , which was slightly higher than the present study. However, stroke is uncommon in TAVR treatment for AR. Mainly due to a lack of valve calci cation and the simplicity and reliability of the THV implantation in patients.

Hemodynamics.
In the present study, the low mean aortic valve gradient of 8.2±3.2 at 30-day, 10.2±4.1 mmHg at 1-year is consistent with previous reports of the same valve 9 . The study result is similar to the 1-month mean valve gradients of other valves (7.9±4.0 mmHg and 7.7±5.1 mmHg reported with the Jenevalve 6 and Direct Flow Medical valve 29 ), although the results provided little information on the effects of the stent on AR hemodynamics. All patients had mild or less PVL at 1-year in the study, with no patients having moderate or severe PVL. These results compare favorably to a recent study that residual moderate or severe aortic regurgitation rate after the procedure was high as 9.2% in other studies 24 . Several characteristics include enhanced positioning accuracy, controlled and anatomically correct implantation and improved sealing even in eccentric annular calci cations. The low rates of moderate to severe PVL might be a contributing factor in low all-cause mortality rate at 1-year. However, with the low mortality rate at 1-year compared to patients with AS, patients with AR were younger and faced the risk of bioprosthetic valve deterioration and the need for reintervention.

Pacemakers
In the rst year, a PPM was implanted in 8.9% of patients. Ten patients received PPM at 30-day, although it was lower than reported with the self-expanding CoreValve 6, 7 and higher than reported from the JeneValve registry 16,26 . AR usually have a larger annulus and lack calci cation. These reasons may have contributed to require a deeper depth of implantation and larger size of prosthesis. There is some evidence that THV deeper position in the LVOT is independently associated with a higher PPMI rate and the larger prosthesis size, valve oversizing is also relevant to risk factors for pacemakers 30 .
Study limitations: Data were obtained in a non-randomized fashion, with the lack of comparative arms of patients with severe AR treated by surgery or medical therapy alone. Also, the sample size was relatively small, and the results were single-center collected, potentially introducing selection bias.
This single-center study reported the outcomes of TAVR in treating pure AR. Our results have demonstrated a favorable 1-year survival rate and symptomatic relieving bene ts of TAVR in AR patients.

Conclusion
The 1-year results of the single-center study support the safety and effectiveness of the J-Valve ® in treating the elderly, high-risk patients with severe AR. Low all-cause mortality, major stroke and no moderate and severe PVL were observed. J-Valve has the advantages of self-position and anatomically correct implantation. However, the rate of PPMI was slightly high. More clinical studies and longer-term follow-up are needed to validate these promising ndings further.  Values are mean±SD or n (%).SD, Standard deviation; Post operation In-hospital stay, (days) , Median (QL-QU) 6(5-8) Values are mean n (%), or mean± SD/Median (QL-QU). SD, Standard deviation; Values are mean±SD, or n (%).SD, Standard deviation; Endocarditis 0 0 All-cause mortality, n (%) 0 10(7.4)