Author + information
- Ron Waksman, MD and
- Augusto D. Pichard, MD∗ ()
- ↵∗Reprint requests and correspondence:
Dr. Augusto D. Pichard, Section of Interventional Cardiology, MedStar Washington Hospital Center, 110 Irving Street, Washington, DC 20010.
“Man must be prepared for every event of life, for there is nothing that is durable.”
Until recently, the standard of care for patients with severe aortic stenosis (SAS) was surgical aortic valve replacement (SAVR). There is no doubt that SAVR outcomes have been extremely good across broad populations and have resulted in excellent procedural outcomes, including low complication rates and acceptable durability of the valve.
Transcatheter aortic valve replacement (TAVR) was introduced >14 years ago and has since been a continuation of technology iteration and improvement, patient selection, and implantation techniques (2). Today, with >125,000 procedures performed in 750 global centers and a substantial amount of emerging data, TAVR is challenging the exclusivity of SAVR for the treatment of patients with SAS. Initially, the efficacy and safety of TAVR were demonstrated in inoperable patients, also known as those at extreme risk for surgery. For this population, TAVR produced reductions in overall mortality (compared with medical therapy or balloon valvuloplasty) with both balloon-expandable and self-expanding valves (3,4).
As a result of these studies, the U.S. Food and Drug Administration granted approval for the marketing of TAVR, and it has become the standard of care in patients who are at extreme risk for surgery. To become a mainstream therapy for patients with aortic stenosis, 2 important questions need to be addressed: 1) will TAVR be noninferior to surgery for the lower risk population? and 2) will the durability of TAVR match SAVR?
The next step was to examine the performance of TAVR versus SAVR among patients at increased risk for surgery. Both balloon-expandable and self-expanding TAVR demonstrated noninferiority to surgery for this patient population (5). Interestingly, the self-expanding valve also showed superiority in survival by an absolute 4.9 percentage points over surgery (6). These results were received with excitement, surprise, and some skepticism because there were a few missing endpoints in the TAVR group, and superiority would not have been established had the conventional 1-sided alpha level of 0.025 been used (7). In addition, there were other trial conduct issues that introduced selection bias and potentially confounded the analysis.
In this issue of the Journal, Reardon et al. (8) present the 2-year results of the study and show not only persistence of the 1-year results but also an increase in the reduction from 4.9 to 6.5 percentage points in mortality in favor of the self-expanding valve compared with the surgical arm. Thus, the reduction in overall mortality, which was of uncertain significance in the 1-year data, is now confirmed with a p value of 0.04. Similar continued divergence was seen in the stroke rates of 6.5% at 1 year to 8.9% at 2 years. The question is whether the differences between years 1 and 2 are statistically significant. A landmark analysis could support whether there is a continuation of improvement in the long-term outcome in the TAVR group. There are several questions that arise from this paper. 1) What is the mechanism for the superiority of the self-expanding valve for mortality reduction and for the improvement from the first to the second year? 2) Are these results consistent regardless of the access site (transfemoral vs. transaortic)? 3) Are these results specific to self-expanding valves, or can they be generalized as a class effect for balloon-expandable valves? We address these 3 points in the following discussion.
Reardon et al. (8) suggest that higher rates of bleeding, transfusion, acute kidney injury at the time of surgery, and higher atrial fibrillation rates in patients treated with surgery may worsen long-term valve performance and can explain the superiority of TAVR over SAVR. Although these arguments are appealing, they are not entirely supported by published reports. For example, hemodynamic differences, such as those presented in the study, are clinically meaningless. In the PARTNER (Placement of Aortic Transcatheter Valve) trial, aortic valve area or gradients post-procedure did not affect clinical outcome, although they were statistically better in the TAVR group. In contrast, paravalvular regurgitation post-TAVR, an established predictor for mortality (9), was higher in the TAVR group in the self-expanding study (6.1% vs. 0.6% moderate to severe, or 36% vs. 7.8% mild, moderate, and severe); however, it did not influence mortality in the present study. Although the data show that paravalvular leaks decrease with time after self-expanding valves, we look forward to the authors publishing the analysis for all-cause mortality stratified according to paravalvular regurgitation at discharge. In addition, the high pacemaker rates and vascular complications (which were also higher in the TAVR group) cannot support improvement in the mortality or stroke rates over 2 years of follow-up. Therefore, the mechanism for superiority of TAVR over SAVR in this analysis remains uncertain.
Because approximately 17% of the cases in the TAVR arm were performed via a noniliofemoral approach (8), the statements of superiority should be carefully reviewed for the TAVR alternative access. In PARTNER 1A, when nearly 30% of the patients were assigned to the transapical approach, SAVR was superior to the transapical access with the balloon-expandable valve for the primary endpoint of the study. The use of alternative access is gradually decreasing with the introduction of low-profile TAVR devices. The evidence for noninferiority of TAVR via alternative access to conventional SAVR is lacking, as are data for the optimal alternative access (e.g., transapical, transaortic, subclavian).
Although Reardon et al. (8) present data limited to the self-expanding valve versus surgery, it does raise the questions of whether these results can be viewed as a class effect and whether they also apply to the balloon-expandable valves. In the PARTNER trial, the patients at high risk for surgery were randomized to receive balloon-expandable valves, and an early benefit of these valves over surgery was found. This survival benefit, however, was lost early, and between 1 and 2 years, the survival curves were similar. Nevertheless, comparing valves across different studies would not be appropriate because of the differences in certain factors, including the study population profile and choice of access. To support the class effect argument, a recent analysis comparing balloon-expandable valves versus SAVR found that for the patients with no or trace paravalvular regurgitation, there was a reduction in mortality in the balloon-expandable valves with transfemoral access compared with SAVR over a 5-year follow-up (45.2% vs. 60.9%), emphasizing the need to eliminate paravalvular regurgitation post-procedure (10). However, not all valves are equal, and to support a class effect, it would require a well-powered randomized trial either comparing the valves head-to-head or against surgery.
Clearly, there are identified populations that are doing better with TAVR versus SAVR. In the PARTNER studies, female sex, patients with smaller annular size, and patients with low-flow low gradient had better outcomes with TAVR. In the present study (8), patients who were <85 years of age, had a Society of Thoracic Surgeons score ≤7%, and those with normal left ventricular function, without hypertension, had a better survival if they underwent TAVR compared with SAVR. This finding is promising and may hint at what we should expect with the results of TAVR versus SAVR in the intermediate-risk populations. If this trend continues and durability of the valve equates with surgery, it will affirm TAVR as the preferred intervention for younger and healthier patients with aortic stenosis. Even if TAVR is noninferior only to SAVR, it would be sufficient to have TAVR as the default strategy for patients with SAS.
The most important question is whether TAVR should become the default therapy for patients with SAS who are at an increased surgical risk and beyond, depending on the durability of the valve. The 2-year data from the study by Reardon et al. (8) are a strong signal to support this approach for the patient with increased surgical risk. The patient’s choice is for the less invasive option, early ambulation, fewer complications, improved quality of life, and durable outcome. The totality of the 5-year data with balloon-expandable valves is sparse because of the advanced age and comorbidities of patients who enrolled in the studies. Thus far, the signals are encouraging regarding the functionality of the valve for those who reached the 5-year follow-up and beyond (11). The present study and others confirm that the intermediate-term results are encouraging regarding the functional performance of TAVR compared with surgery in the high-risk population.
Looking forward, the message is clear: TAVR via femoral access should be considered as the first-line therapy for patients with SAS who are at an increased risk for surgery. SAVR should be considered only for patients who are not suitable for TAVR or when the data are not available yet (intermediate- and low-risk PARTNER 2A and SURTAVI [Surgical Replacement and Transcatheter Aortic Valve Implantation]). Published reports have already noted the noninferior results of TAVR versus SAVR for lower risk populations (12). The majority of surgeons are currently embracing TAVR and converting SAVR cases to TAVR as part of the heart team approach, making the possibility that TAVR will become mainstream therapy a reality (13).
With the rapid accumulation of data, it is time for the societies to update and upgrade their guidelines and equate them to the surgery guidelines for the high-risk patient population (14). Insurers should also recognize the totality of the evidence that supports TAVR and upgrade reimbursements to be in synchronization with labeling by the U.S. Food and Drug Administration. Cost should not be an obstacle or prohibit patients with SAS from having the best evidence-based medicine therapy available.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Dr. Waksman has received personal fees from Biotronik, Medtronic, and Abbott Vascular; grants and personal fees from AstraZeneca, Boston Scientific, and Biosensors International; and grants from The Medicines Company and Edwards Lifesciences. Dr. Pichard has reported that he has no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- ↵(1917) Forty Thousand Quotations: Prose and Poetical, comp. by Charles Noel Douglas (Halcyon House, New York, NY).
- Cribier A.,
- Eltchaninoff H.,
- Tron C.,
- et al.
- Popma J.J.,
- Adams D.H.,
- Reardon M.J.,
- et al.
- Reardon M.J.,
- Adams D.H.,
- Kleiman N.S.,
- et al.
- Genereux P.,
- Head S.J.,
- Hahn R.,
- et al.
- Mack M.J.,
- Leon M.B.,
- Smith C.R.,
- et al.,
- for the PARTNER 1 Trial Investigators
- Kapadia S.R.,
- Leon M.B.,
- Makkar R.R.,
- et al.,
- for the PARTNER Trial Investigators
- Thyregod H.G.,
- Steinbrüchel D.A.,
- Ihlemann N.,
- et al.
- Nishimura R.A.,
- Otto C.M.,
- Bonow R.O.,
- et al.