Aortic stenosis is the most common valvular heart disease. In the last decade, transcatheter aortic valve replacement (TAVR) has replaced surgical aortic valve replacement as the most common treatment for symptomatic severe aortic stenosis in the United States (1-5). The safety and efficacy of TAVR has been established in all risk groups within the manufacturer guidelines for annular size; however, the options for those patients having an annular size outside of the guidelines remain unclear.

The only third-generation transcatheter heart valves indicated for large aortic annuli (area 575-683 mm2; perimeter 85.0-94.2 mm) are the 29-mm SAPIEN 3 (Edwards Lifesciences, Irvine, Calif.) and the Evolut R, Evolut PRO, Evolut PRO+, and Evolut FX (Medtronic, Minneapolis, Minn.). Use of these valves in extra-large annuli (area ≥683 mm2; perimeter ≥94.2 mm) is considered off-label. To our knowledge, only 2 single-center case series, each with a limited number of off-label patients, have explored outcomes in this population (6,7).

The current study was designed to report our center’s experience with the safety and efficacy of using self-expanding valves (34-mm Evolut R, Evolut PRO, Evolut PRO+, and Evolut FX) in patients with off-label aortic annulus size (cases) in comparison with using these 34-mm TAVR valves in patients with an on-label aortic annulus size (controls). All patients in the off-label group were rejected for surgery based on their assessment, which included STS score and multiple other variables, and recommended by the patient care team (cardiothoracic surgeon, interventional cardiologist, and imagers) for TAVR as the only available therapeutic option.

We hypothesized that TAVR would be a safe and effective treatment in these high-risk, off-label patients and would not result in a greater burden of postoperative paravalvular leak (PVL), permanent pacemaker (PPM) implantation, or other serious complications.

This study was approved by the Advocate Aurora Health Institutional Review Board and considered exempt status. A waiver of informed consent was granted. The research was conducted in accordance with the Declaration of Helsinki.

Study Setting. Data were collected retrospectively from the electronic medical records of all patients who had 6 TAVR with a 34-mm Evolut R, Evolut PRO, Evolut PRO+, or Evolut FX valve through the Structural Valve Program at Aurora St. Luke’s Medical Center in Milwaukee, Wisconsin, from October 2016 to June 2022. Patient demographic information, echocardiography and computed tomography (CT) findings, procedural characteristics, and clinical outcomes data entered into the medical record prospectively as part of the STS/ACC TVT (Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy) Registry were collected. After exclusion of valve-in-valve patients and those without a CT scan of the heart, 530 patients (81±8.3 years of age, 88% male) were included in the study. A total of 494 patients were in the on-label group (controls) and 36 in the off-label group (cases).

Critical aortic valve stenosis was defined as an aortic valve area of <0.6 cm2. All definitions used were those recorded in the STS/ACC TVT Registry Adverse Event Definitions v2.1 and VARC-3 criteria (8). Aortic valve annular measurements were obtained and measured with electrocardiogram-gated cardiac CT. Discharge status was determined at the moment each patient left the hospital grounds. Post-TAVR aortic insufficiency was classified as none, trace/trivial, mild, moderate, or severe by a transthoracic echocardiogram read on the first postoperative day by a TAVR-dedicated, level III echocardiologist. For outcome assessment, PVL was characterized as none, trace, mild, moderate, or severe, with only moderate or severe PVL considered clinically relevant for analysis. Valve function and performance was examined by obtaining the pre- and post-TAVR echocardiograms for each patient.

The off-label group (cases) was the group of interest in this study. In order to compare their outcomes, a control group was created by selecting all patients in our database who underwent TAVR with the same size valves (Evolut 34 R, Evolut PRO, Evolut PRO+, or Evolut FX) during the same time period at the same center but were on-label, meaning that their annular size was in accordance with the manufacturer guidelines (annular perimeter <94.2). Additionally, the control group included patients in all risk categories based on STS score, except for those who were classified as surgically prohibitive risk. This creates a bias in the study design, favoring the control group to have a better outcome. We deemed this acceptable because the study aimed not to show superior outcomes in cases, but to determine whether TAVR can offer the case group reasonable outcomes compared with controls, given that cases have no other treatment options.

Preprocedural electrocardiogram-gated cardiac CT was used to obtain annular and anatomic measurements for study analysis.

Definitions. VARC-3 definitions were utilized for complications (Table A.1) (8).

Statistical Analysis. Patient data were collected from the electronic medical record and organized into one dataset. Continuous variables were represented as mean ± standard deviation. Variables were compared using a one-way analysis of variance or Student’s t-test. Nominal data were represented as n (%). Variables were compared using the chi-square test.

Demographics, echocardiographic characteristics, CT characteristics, and clinical outcomes were analyzed and compared between cases and controls. If any continuous variable had a skewed distribution, a Wilcoxon or Kruskal-Wallis test was performed or the variable was converted to nominal and a one-way chi-square approximation test was performed.

Stratum-specific analyses were performed to identify potential confounding variables and effect-modifying variables. The alpha value of 0.05 was used to determine the statistical significance of the data. All statistical analyses were performed using JMP software (SAS Institute, Cary, North Carolina).

The demographic and clinical characteristics of the patient population are presented in Table 1. The off-label group was younger (77 vs 81 years; P=0.004), taller (178 vs 173 cm; P=0.001), and heavier (100 vs 90 kg; P=0.003), had a higher prevalence of male patients (100% vs 87%; P=0.02), and had a lower incidence of peripheral artery disease (11.1% vs 26.6%; P=0.04). Interestingly, there was no statistical difference in the STS risk score between groups (off-label 4.15 vs on-label 4.10; P=0.9).

An analysis of the preoperative echocardiographic parameters (Supplemental Table 1) did not demonstrate any statistical differences in left ventricular ejection fraction, right ventricular systolic pressure, left atrial volume index, aortic valve area, aortic valve peak gradient, or rate of critical aortic stenosis between the groups.

Baseline CT characteristics (Supplemental Table 2) demonstrated a greater annular perimeter (97.2 vs 84.7 mm; P≤0.0001), mean sinus of Valsalva diameter (37.8 vs 34.5 mm; P≤0.0001), right coronary artery height (20.2 vs 17.9 mm; P=0.0005), and mean sinus of Valsalva height (18.5 vs 16.8 mm; P=0.005) in the off-label group compared with the on-label group.

Procedural characteristics are shown in Supplemental Table 3. The 34-mm Evolut R was the most used valve (76%), although it trended toward being used less often in off-label cases than on-label controls (61.1% vs 77.7%; P=0.02). The Evolut PRO and PRO+ valves were more commonly used in off-label cases because they are larger valves that have an additional outer skirt (Evolut PRO) or an external pericardial wrap around the valve frame (Evolut PRO+). The type of anesthesia, amount of contrast, and total fluoroscopy time were not significantly different between groups. However, the off-label group had a greater incidence of monitored anesthesia care (92.7% vs 82.9%, p=0.03).

Supplemental Table 4 shows postoperative TAVR valve performance based on echocardiographic parameters. In the off-label cases, the peak velocity (1.93 vs 2.16 m/s; P=0.004) was lower and the aortic valve area larger (2.52 vs 2.14 cm2; P=0.002). The mean aortic valve gradient was lower in the on-label group than the off-label group (7.07 vs 8.1 mmHg; P=0.07), but this difference was not statistically significant.

The off-label group had a significantly higher incidence of multiple post-TAVR complications (Table 2), including PVL (11.1% vs 1.4%; P≤0.0001), unplanned other cardiac surgery or intervention (5.7% vs 0.4%; P=0.0006), device migration (5.6% vs 0%; P≤0.0001), and in-hospital mortality (5.6% vs 0.8%; P=0.0094). Although complications were more common in the off-label cases, the majority of patients in this group had no significant PVL (89%) and the vast majority survived the procedure (94%). Moderate or severe PVL was more frequent in the off-label group (n=4, 11.1%) than on-label group (n=7, 1.4%). The rates of aortic valve reintervention, PPM implantation, and vascular complications were found to be statistically similar in both groups (0% vs 0.4%; P=0.70; 8.3% vs 13.6%; P=0.37; 5.7% vs 4.3%; P=0.71, respectively).

Of the 530 patients who underwent TAVR, 6 died while in the hospital; of these deaths, 2 were in the off-label group and 4 in the on-label group (Supplemental Table 5). In the off-label group, the deaths were caused by valve migration (n=1) and aortic perforation (n=1). Causes of death in the on-label group were aortic perforation (n=1), pulseless electrical activity arrest after 4 days (n=1), cardiac standstill during TAVR (n=1), and renal failure/hospice (n=1).

Table 1: Demographic and clinical characteristics of population for Evolut 34R, Evolut 34 PRO, Evolut PRO+, and Evolut FX patients

Data are presented as mean±SD or n (%).

P<0.05 is considered significant.

^aMissing 74 patients due to uncertainty about morphology.

BAV, bicuspid aortic valve; LBBB, left bundle branch block; PAD, peripheral artery disease; PCI, percutaneous coronary intervention; RBBB, right bundle branch block; STS, Society of Thoracic Surgeons.

Table 2: Post-TAVR complications

Data are presented as n (%).

P < 0.05 is considered statistically significant.

ICD, implantable cardioverter-defibrillator; LVOT, left ventricular outflow tract; TAVR, transcatheter aortic valve replacement; THV, transcatheter heart valve.

Table 3: TAVR in large and extra-large aortic annuli

ER, self-expanding valve; PPM, permanent pacemaker; PVL, paravalvular leak; S3; balloon-expandable valve; TAVR, transcatheter aortic valve replacement.

SUPPLEMENTAL MATERIAL

Supplemental Table 1: Echocardiographic characteristics of population

Data presented as mean±SD or n (%).

P<0.05 is considered statistically significant.

^aMissing data for a sample of patients.

^bCritical aortic valve stenosis = aortic valve area < 0.6 cm².

AV, aortic valve; LAVI, left atrial volume index; LVEF, left ventricular ejection fraction; LVIDd, left ventricular internal diastolic dimension; RVSP, right ventricular systolic pressure.

Supplemental Table 2: Computed tomography characteristics of population

Data are presented as mean±SD.

^aMissing 1 patient from sample

P<0.05 is considered statistically significant.

Supplemental Table 3: Procedural characteristics

Data are presented as mean±SD or n (%).

^aMissing 1 patient from dataset.

P<0.05 is considered statistically significant.

Supplemental Table 4: Post-TAVR valve performance

Data are presented as mean±SD.

P<0.05 is considered statistically significant.

^aMissing 2 patients from dataset.

^bMissing 3 patients from dataset.

AV, aortic valve; TAVR, transcatheter aortic valve replacement.

Supplemental Table 5: Mortality data

BAV, bicuspid aortic valve; CVA, cerebrovascular accident; PEA, pulseless electrical activity; PVL, paravalvular leak; SAVR, surgical aortic valve replacement; STS, Society of Thoracic Surgeons; TAVR, transcatheter aortic valve replacement; VSD, ventricular septal defect.

This study represents the largest study to-date to evaluate TAVR in patients with off-label, oversized annuli who have surgically prohibitive risk, including 36 such patients within a total cohort of 530. Two other studies reported experiences with extra-large annuli, but neither of these studies were carried out with patients with surgically prohibitive risk. Our cohort had larger annuli, with a mean perimeter of 97.2 ± 3.8 mm, compared with 94.5-97.9 mm in Armijo et al. (6) and 96.7 ± 2.7 mm in Sengupta et al. (7). Despite this, comparable outcomes were achieved using self-expanding valves. Specifically, a device success rate of 94.4% with 0% in-hospital stroke, 0% valve embolization, and 0% need for aortic valve reintervention was observed.

Baseline Characteristics. Table 1, Supplemental Table 1, and Supplemental Table 2 show that the off-label group had a greater prevalence of male sex, greater height, greater weight, and larger aortic measurements (annular perimeter, sinus of Valsalva diameter and height, and right coronary artery height) on CT, all likely reflective of larger aortic annulus size. Although STS score was similar in cases and controls (4.2 vs. 4.1), the mortality rate was much higher in cases (5.6% vs. 0.8%), underscoring the inherent limitation of STS score as compared with a surgeon’s clinical judgement.

Procedural Characteristics: Valve Choice, Contrast Volume, Fluoroscopy Time. The procedural characteristics (Supplemental Table 3) show that controls were more likely to receive an Evolut R (34 mm) valve while cases received more of the newer valves (Evolut PRO+ and Evolut FX), likely because the newer valves have a pericardial skirt to confer a larger annular area to the cases. The off-label case group more frequently needed monitored anesthesia care, more contrast volume, and longer fluoroscopy time, likely related to the higher comorbidities of the cases.

Post-TAVR Valve Performance. Post-TAVR valve performance (Supplemental Table 4) was inferior in cases with higher peak velocity, higher gradients, and lower valve area, but these differences were not large enough to be clinically meaningful.

Complications of Off-Label Vs On-Label Aortic Annulus Size in TAVR. Tables 2 and 3 summarize and compare the current study with 2 other studies (6,7) performed in patients with large and extra-large annuli undergoing TAVR. As expected from the study designs, cases in the current study had higher mortality and PVL rates compared with the controls as well as compared with those in other studies. However, other complications were similar, supporting the notion that TAVR with a 34-mm valve is a viable option in these patients who have no other options. Both deaths in the off-label group resulted from mechanical complications (valve migration and aortic perforation), likely related to technical challenges in the cases due to extra-large annuli and comorbidities, respectively. The technical challenges in treating oversized annuli in sicker patients cannot be overemphasized. They demand use of newer valves with a pericardial skirt and CT evaluation of the aorta as extreme tortuosity increases the risk of perforation.

Limitations. This study has several limitations that should be considered when interpreting the results. First, the lack of mid- and long-term follow-up data limited our ability to assess durability and outcomes over extended periods. Additionally, our center predominantly utilizes self-expanding valves, and we have limited experience with balloon-expandable valves. Although this limits the generalizability of our findings to centers that primarily use self-expanding valves, it also serves as a strength of the study, as we present the largest single-center experience with self-expanding valves. It is important to note that self-expanding and balloon-expandable valves may exhibit different behaviors, and our findings provide valuable insights specific to the self-expanding valve population.

Another limitation is the retrospective nature of this study, which is susceptible to selection bias and missing data. The results may be influenced by the specific inclusion criteria and patient characteristics at our center, limiting the external validity to other populations. Further, the relatively small sample size and single-center design may reduce the power and generalizability of our findings. Although we have taken steps to mitigate potential biases, such as using robust statistical methods, the possibility remains that unmeasured confounders affected the outcomes.

Lastly, we currently do not have a direct comparison of Evolut TAVR with surgical outcomes or other TAVR devices as our study focused exclusively on patients deemed to have prohibitive surgical risk who were managed with optimized medical therapy and for whom TAVR represented the last treatment option. We hope to address this in future research as additional data become available. Despite these limitations, this study provides important insights into the outcomes of self-expanding valves in a large cohort, and we aim to expand the scope of our analysis in future investigations.

This study is, to our knowledge, the largest study of off-label Evolut TAVR valve implantation in patients who had severe aortic stenosis and oversized aortic annuli per the manufacturer guidelines and who had been turned down for surgical intervention. The results demonstrate that TAVR is relatively safe and effective in this off-label population that otherwise has a 50% mortality rate at 1 year. Based on this study, all patients who meet these criteria should be considered appropriate candidates for TAVR with an Evolut valve, as this represents a viable therapeutic option with meaningful clinical benefit for patients otherwise excluded from surgical intervention.

Table A.1. Definitions of complications

1. Leon MB, Smith CR, Mack M, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med 2010;363:1597-1607
2. Smith CR, Leon MB, Mack MJ, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med 2011;364:2187-2198
3. Kodali SK, Williams MR, Smith CR, et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med 2012;366:1686-1695
4. Adams DH, Popma JJ, Reardon MJ, et al. Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med 2014;370:1790-1798
5. Popma JJ, Adams DH, Reardon MJ, et al. Transcatheter aortic valve replacement using a self-expanding bioprosthesis in patients with severe aortic stenosis at extreme risk for surgery. J Am Coll Cardiol 2014;63:1972-1981
6. Armijo G, Tang GHL, Kooistra N, et al. Third-generation balloon and self-expandable valves for aortic stenosis in large and extra-large aortic annuli from the TAVR-LARGE Registry. Circ Cardiovasc Interv 2020;13:e009047
7. Sengupta A, Zaid S, Kamioka N, et al. Mid-term outcomes of transcatheter aortic valve replacement in extremely large annuli with Edwards SAPIEN 3 valve. JACC Cardiovasc Interv 2020;13:210-216
8. STS/ACC TVT Registry. Transcatheter Aortic Valve Replacement (TAVR) v3 Data Collection Form. Available at https://www.ncdr.com/WebNCDR/docs/default-source/tvt-public-page-documents/tvt-v3-0-tavr-11-18-2021sf.pdf?sfvrsn=96c0d69f_4. Accessed August 8, 2024..