What is the role of transcatheter aortic-valve replacement (TAVR) in patients with severe, symptomatic aortic stenosis?
Alain Cribier performed the first human TAVR April 2002 in Rouen, France. This first TAVR patient was a 57-year-old man with severe AS who had been deemed ineligible for SAVR due to hemodynamic instability and significant comorbidities. The patient survived the procedure with no subsequent episodes of acute, decompensated heart failure but died 17 weeks post-TAVR due to complications from a non-healing skin/soft-tissue infection in the setting of limb ischemia. In the two decades that followed, the Placement of AoRtic TraNscathetER valves (PARTNER) series of randomized controlled trials examined all-cause mortality in patients undergoing SAVR vs. TAVR, first in high-, then intermediate-, and finally in low-risk surgical patients. These studies all utilized the Edwards SAPIEN heart-valve system.
Mr. M Jagger, a 75-year-old man and renowned rock-and-roll artist, was planning an extensive North American tour with his band that was postponed when he developed progressive shortness of breath and bothersome swelling in his legs. He underwent several tests including a TTE, which demonstrated an LV ejection fraction of 40% and severe aortic stenosis (AS). He and his doctor discussed surgical aortic valve replacement (SAVR) and transcatheter aortic-valve replacement (TAVR) as treatment options. Mr. Jagger’s main priority is getting back on the road with his band as soon as possible. While he knows You Can’t Always Get What You Want, he wonders if the less invasive TAVR will help him get what he needs—that is, rolling with The Stones on tour.
By the end of this post, you should have a basic understanding of a few of these trials and how you might counsel M. Jagger.
The Trials
PARTNER-1 (Cohort B, 2010)
In patients with severe AS and PROHIBITIVE SURGICAL RISK, TAVR was found to to reduce mortality when compared to standard (medical) care.
Population (N=358)
- Patients with severe AS and NYHA II-IV heart failure who were poor surgical candidates (STS risk score ≥ 10%)
Intervention
- TAVR (n= 179)
- Standard therapy (n=179)
Outcomes
- Significant decrease in all-cause mortality at 1 year with TAVR (30.7% vs. 49.7%)
- Significant decrease in all-cause mortality or time to first repeat hospitalization for valve, or procedure-related deterioration at 1 year with TAVR (42.5% vs. 70.4%)
- TAVR group had significantly higher rates of stroke and vascular complications at 1 year (secondary outcome)
Criticism
- Some relevant groups excluded (CAD, PVD, EF less than 20%)
- Standard therapy involved aortic valvuloplasty which carries a class III recommendation (no benefit)
- TAVR benefit may reflect harm from valvuloplasty
PARTNER-1 (Cohort A, 2011)
In patients with severe AS and HIGH SURGICAL RISK, TAVR was found to be non-inferior to SAVR.
Population (N=699)
- Patients with severe AS and NYHA II-IV heart failure who were high-risk surgical candidates (STS risk score ≥10%)
Intervention
- TAVR (n=348)
- SAVR (n=351)
Outcomes
- Met statistical endpoint for non-inferiority of TAVR vs. SAVR with regards to all-cause mortality at 1 year (24.2% vs. 26.8%)
- Vascular complications were higher in the TAVR group at 1 year (secondary outcome)
- Major bleeding at 1 year was higher in the SAVR group (secondary outcome)
- Five-year results indicate that hemodynamic and clinical outcomes remained similar between groups
Criticism
- Some relevant groups excluded (CAD, PVD, EF less than 20%)
- Nearly 5% crossover to SAVR may have biased results in favor of TAVR
PARTNER-2 (2016)
In patients with severe AS and INTERMEDIATE SURGICAL RISK, TAVR was found to be non-inferior to SAVR.
Population (N = 2032)
- Patients with severe AS and NYHA II-IV heart failure who were intermediate-risk surgical candidates (STS score ≥4%)
Intervention
- TAVR (n=1011)
- SAVR (n=1021)
Outcomes
- Met statistical endpoint for non-inferiority of TAVR vs. SAVR with regards to all-cause mortality or disabling stroke at 2 years (19.3% v 21.1%)
- TAVR caused more major vascular complications at 2 years (secondary outcome)
- SAVR caused more life threatening or disabling bleeds at 2 years (secondary outcome)
Criticism
- Limited follow-up
- Some relevant groups excluded (CAD, PVD, EF less than 20%)
- Greater than 10% study population with STS risk score above 8% which may bias results towards TAVR
PARTNER-3 (2019)
In patients with severe AS and LOW SURGICAL RISK, TAVR was found to be both non-inferior and superior to SAVR.
Population (N = 1,000)
- Patients with severe AS and NYHA II-IV heart failure who were low-risk surgical candidates (STS risk score ≤ 4%)
Intervention
- TAVR (n=503)
- SAVR (n=497)
Outcomes
- With regards to a composite of all-cause mortality, stroke, or rehospitalization at 1 year TAVR met statistical criteria for both non-inferiority and superiority (8.5% [TAVR] vs. 15.1% [SAVR])
- Significantly more strokes, major bleeds, and new onset atrial fibrillation in SAVR group (secondary outcome)
- Significantly more patients discharged to home in TAVR group (secondary outcome)
Criticism
- Limited follow-up
- All patients in the TAVR group were transfemoral access candidates, which is known to be associated with improved outcomes when compared to other access sites
- TAVR associated with considerably more paravalvular leak at 1 year than SAVR
- Some relevant groups excluded (bicuspid AV)
The Bottom Line
The PARTNER series examined patients with a broad spectrum of surgical risk and demonstrated that TAVR is better than or as good as SAVR. In 2019 the FDA expanded the indication for TAVR to include low-risk surgical candidates in addition to intermediate- and high-risk patients. Although an ongoing criticism of all PARTNER studies is the lack of long-term follow-up in TAVR patients, the 5 year data has now been released for patients with prohibitive, high, and intermediate surgical risk. All three have demonstrated no significant difference in all-cause mortality, functional outcomes, and valve hemodynamics. Future studies are needed to define the longer-term risks and benefits of this procedure.
Expert’s Opinion
Olcay Aksoy, MD, FACC
Assistant Professor, Interventional Cardiology
Cardiovascular Medicine
It wasn’t long ago that ~40% of patients were told that they had no treatment options for their aortic stenosis. Many of these patients were at most offered balloon aortic valvuloplasty (BAV) for symptom relief—which is partially successful but with high risk. As technology evolved and stents became widespread, the idea of mounting valves inside of the stents and delivering them into native diseased valves became a reality. These ideas were then tested in several rigorous, robust clinical trials including the PARTNER studies listed above—which paved the way for approval by the FDA. TAVR has since become the preferred treatment modality for most patients.
After the PARTNER 1 study, a majority of those patients who were once told that they had no treatment options are now routinely being treated with Transcatheter Aortic Valve Replacement (TAVR) procedures with significant survival benefit as compared to medical therapy. Subsequent trials have compared TAVR to the surgical treatment modality—and have shown consistent findings with at least comparable if not better outcomes with low rates of mortality and stroke. As discussed above, PARTNER 2 and PARTNER 3 trials have since led to the approval of TAVR for intermediate and low risk patients and we now routinely perform these procedures for a majority of patients.
We should note that the two treatment modalities are inherently different. The surgical approach has been time tested and been very successful. The procedure involves removal of the diseased valve and suturing down a new valve. Bioprosthetic valves usually last 10-15 years and mechanical valves can last a life time—though warfarin is used. STS score is used to estimated 30-day mortality risk and guides us in our decision making.
TAVR, on the other hand, involves delivery of the valve-in-stent via a catheter and the stent is subsequently deployed into the diseased valve. The valve is delivered through catheters mostly from the groin (as new designs are developed, greater than 95% of valves are placed from the groin with other access sites including axillary artery, carotid artery, inferior vena cava-aorta approach, and left ventricular apical access). Most procedures last less than 1 hour and are done with conscious sedation with ambulation of the patient the same day.
Both procedures have pro’s and con’s. While surgical approach has longer term data available, its invasive nature and data from the randomized trials make this approach less appealing in many cases. TAVR has limited long term data but enjoys excellent outcomes and very quick recovery times with the majority of patients going home the next day post-procedure. Risk of complications remains low with essentially immediate symptom improvement. While quicker recovery times and convenience of TAVR procedure makes it more attractive, one has to continually monitor the long term data–and in fact, a 5 year interim analysis of the PARTNER 2 data showed that surgical approach may be associated with better outcomes especially in non-transfemoral cohorts with curves separating for lower rates of death or disabling strokes in favor of surgical cohort (N Engl J Med 2020; 382:799-809, Feb 27, 2020). As such, we need to be diligent about monitoring and analyzing the long term data going forward.
When evaluating a patient, all cases are reviewed by heart teams that include cardiologists and cardiovascular surgeons. The variables examined include age, expected future survival, surgical candidacy, quality of life, and anatomic factors such as calcification of the valve, coronary occlusion risk, presence of aortic aneurysm, and concomitant coronary disease or valvular disease.
It is with great pleasure to report that we can offer most of our patients the TAVR procedure with low risk and excellent outcomes. While the long term data is still pending, the procedure has taken hold and enjoys significant success. Given its penetration and excellent outcomes, many other valves are also being developed for mitral, tricuspid, and pulmonic positions, and we look forward to incorporating these minimally invasive options into our practice as we await results from more randomized trials.
No conflicts of interest.
Authors of this edition
Ethan Greenblatt, MD
Chief Resident, UCLA Ronald Reagan Medical Center
Shire Beach, MD
PGY-3, UCLA Ronald Reagan Medical Center