September 18, 2018
6 min read

The Evolution of TAVR From a High-Risk Procedure to Standard of Care

More patients lower on the risk spectrum are receiving TAVR, but surgical AVR should not be counted out.

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

The evidence base for transcatheter aortic valve replacement in the United States is now extensive. Randomized controlled trials have been conducted in more than 7,000 U.S. patients and numerous publications in journals such as The Lancet and The New England Journal of Medicine have shown it is just as effective as surgical AVR in various populations.

For both balloon-expandable valves (Sapien family of products, Edwards Lifesciences) and self-expanding valves (CoreValve family of products, Medtronic), the evidence shows that for patients with severe aortic stenosis, they are superior to medical therapy in inoperable patients and they are similar to surgical AVR in high-risk and intermediate-risk patients. Trials of both valve types vs. surgery in low-risk patients have completed enrollment, with results anticipated to be available in early 2019.

While it is easy to predict that some day TAVR will become predominant over surgical AVR, several factors may favor surgical AVR in certain types of patients for the foreseeable future.

Michael J. Mack

Extensive Evidence

The evidence for the efficacy of TAVR was strong enough to prompt in 2017 an update to the 2014 American College of Cardiology/American Heart Association guideline for management of patients with valvular heart disease. TAVR is now a class I recommendation for patients at prohibitive or high surgical risk and a class IIa recommendation for patients at intermediate surgical risk. At the moment, only surgical AVR is recommended for patients at low surgical risk, but that may change depending on the outcomes of the low-risk TAVR trials.

As the evidence for TAVR has grown, so has its usage in the U.S. According to the Society of Thoracic Surgeons/ACC Transcatheter Valve Therapy (TVT) registry database, the number of U.S. sites performing TAVR rose from 156 in 2012 to 587 in August 2018. Similarly, commercial U.S. TAVR procedures have risen steadily, from less than 5,000 in 2012 to nearly 50,000 in 2017. In fact, in 2017, for the first time, TAVR volume eclipsed surgical AVR volume, even when isolated surgical AVR was considered together with combination AVR and CABG procedures (see Graph).

As TAVR has evolved into a procedure for patients further down the spectrum of risk, it is no longer considered exclusively for the very elderly. Indeed, the mean age of U.S. patients who underwent TAVR fell from 82 years in 2014 to 80 years in 2017.

TAVR’s progression from a high-risk procedure into the standard of care is reflected in a number of statistics. In 2013, according to the TVT registry, total length of stay was 7 days and post-TAVR length of stay was 6 days, but by 2017, that had dropped to 3 days and 2 days, respectively. In 2017, 85% of patients were discharged home instead of to a nursing home or rehabilitation facility after TAVR, up from 66.4% in 2012.


Procedural complications have fallen as TAVR volume has grown. In 2012, 4.2% of cases required cardiopulmonary bypass, but that number fell to 0.7% in 2017. Similarly, conversion to open-heart surgery happened 1.4% of the time in 2012 but 0.5% of the time in 2017, and 3.2% of procedures were aborted in 2012 compared with 0.8% in 2017. In conjunction, as evidence has mounted that the transfemoral approach confers the fewest complications, the percentage of TAVR procedures performed via the transfemoral approach has risen over time.

Mortality rates associated with TAVR have dropped steadily. In-hospital mortality fell from 4.11% in 2014 to 1.74% in the first half of 2017, while 30-day mortality dropped from 6.01% to 3.05% during the same period. From linking the TVT registry to CMS databases, it is estimated that 1-year mortality dropped from 21.3% in 2014 to 15.26% in the first half of 2016.

Stroke rates are now consistently at less than 2% following TAVR, although subject to underreporting, and quality of life metrics are encouraging: Nearly 65% of patients improve by at least 20 points in the Kansas City Cardiomyopathy Questionnaire score at 30 days after their procedure.

Patient Selection Crucial

Given all these promising data, it is not surprising that the next frontier of TAVR is in patients at low surgical risk. The first data from the PARTNER 3 trial of balloon-expandable valves in low-risk patients and the EvolutR low-risk trial of self-expanding valves are expected to be made public in the spring of 2019.

These trials will force us to think even harder about patient selection. Assuming they show TAVR is noninferior to surgical AVR in low-risk populations, we will have to consider factors that may mitigate the universal adoption of TAVR and suggest that some patients may remain better candidates for surgical AVR.

Potential factors against TAVR in certain patients will include concerns over the durability of the valves, especially in younger patients; the higher pacemaker implantation rate for TAVR vs. surgical AVR, especially in younger patients; the higher incidence of valve thrombosis for TAVR vs. surgical AVR, which raises the question of long-term anticoagulation; and outcomes in bicuspid aortic valves.


In one study published in The Lancet, reduced leaflet motion was present in 13.4% of transcatheter valves vs. 3.6% of surgical valves, and was more likely to occur in patients not on an anticoagulant.

In another study published in the Journal of the American College of Cardiology, transcatheter aortic valve thrombosis occurred in 7% of patients.

Implantation of new permanent pacemakers has been a concern ever since the advent of TAVR, and even in the newest-generation devices, the rates of new pacemakers at 30 days range from 15% to 20%.

Some of these concerns are being addressed by the ongoing trials in low-risk patients. The PARTNER 3 and EvolutR trials will evaluate a combined 800 patients to determine valve thrombosis via 4-D CT and will follow all 2,000 patients for 10 years to assess the durability of both TAVR and surgical valves.

Before we can consider TAVR to be a universal solution, we must answer a number of questions related to valve thrombosis, including whether it is truly more common with TAVR than with surgical AVR, whether it is a class effect or specific to one or more devices, whether it is clinically significant, whether it causes early structural valve deterioration and which patients need to be anticoagulated, and for how long.

We also need to answer key questions about new pacemaker implantation, including to what degree it is more common with TAVR than surgical AVR, whether it is associated with decreased cardiac function and survival, whether the implant rate is decreasing over time, whether the rate will be less frequent in lower-risk patients and whether there would be more pacemaker-related complications in younger patients.

TAVR in bicuspid aortic valves remains a challenge, as evidenced by one study in JACC which found that compared with TAVR procedures in tricuspid valves, those in bicuspid valves were associated with a lower device success rate. The difference was driven by early-generation valves, so we need to confirm whether this is no longer a concern with the latest-generation valves.

A Guess at Best Candidates

Based on these factors, we can begin to determine which patients are likely to stay with surgical AVR even if TAVR is shown to be noninferior in low-risk patients. These include younger patients where there might be concerns about valve durability and need for permanent pacemaker and who are more likely to have bicuspid valves; patients with multivessel CAD and a high SYNTAX score; patients with rheumatic valve disease; patients at low or intermediate surgical risk with multivalve disease; patients treated at sites that only offer surgical AVR; patients treated at sites that do not have a coordinated heart team; and patients who prefer an open procedure to a transcatheter one.


Therefore, the best candidates for TAVR may include the elderly, the frail, those at intermediate and high risk, those with a small aortic root and those requiring a valve-in-valve procedure (23 mm or greater), while the best candidates for surgical AVR may include those who are younger, those who are at low risk, those who have a bicuspid valve, those with aortopathy, those with CAD and a high SYNTAX score, those needing a smaller redo valve and those with multivalve disease.

While those represent the best guesses at this time, it is never easy to predict the future. In February, researchers reported in Catheterization and Cardiovascular Interventions that in a meta-analysis of randomized trials and propensity-matched observational studies in low-risk patients, TAVR was associated with higher risk for intermediate-term mortality than surgical AVR (RR = 1.45; 95% CI, 1.11-1.89). While it is expected that TAVR volume will continue to grow in the U.S., it is certainly premature to consign surgical AVR to historical status.

Disclosure: Mack reports he is co-principal investigator of the COAPT trial sponsored by Abbott Vascular, is co-principal investigator of the PARTNER 3 trial sponsored by Edwards Lifesciences and is study chair of the APOLLO trial sponsored by Medtronic.