This article is more than 5 years old. Information may no longer be current.
A Relentless Drive to Low Risk
Preprocedure risk level is an important determinant of TAVR success.
Click Here to Manage Email Alerts
Since the first successful transcatheter aortic valve replacement in man by Alain Cribier, MD, in 2002, the procedure has exploded and transformed the field of structural heart disease.
This explosion is based on a strong foundation of randomized data. High-risk randomized trials showed TAVR to be noninferior or superior to surgery. Two intermediate-risk trials showed TAVR to be noninferior to surgery. Two low-risk randomized trials, PARTNER III and Medtronic Transcatheter Aortic Valve Replacement in Low Risk Patients, are now actively recruiting. Understanding this relentless drive to low risk and why it has been so successful to date is integral to understanding the future of structural heart disease.
Encouraging Data
The PARTNER IIA randomized trial randomly assigned 2,032 patients at intermediate risk for surgery to undergo TAVR or surgery. The primary endpoint was all-cause death or disabling stroke at 2 years. This endpoint was reached in 19.3% of the TAVR group and 21.1% of the surgical group overall. The trial was stratified by access site, which was 76.3% transfemoral and 23.7% transthoracic. For the transfemoral group, the 2-year endpoint of all-cause death or disabling stroke for TAVR vs. surgery was 16.8% vs. 20.4% in the intent to treat group (P = .05). The transthoracic group, however, reached this endpoint in 27.7% of those assigned TAVR vs. 23.4% of those assigned surgery, trending toward favoring surgery.
These data led to FDA approval of the Sapien valve system (Edwards Lifesciences) for patients with symptomatic severe aortic stenosis at intermediate risk for surgery. This first randomized trial helped move the authors of the American Heart Association/American College of Cardiology 2017 guidelines to make TAVR a class IIa recommendation.
The SURTAVI trial was recently presented at the 2017 ACC Scientific Session in March and then published in The New England Journal of Medicine. The primary endpoint was the composite of all-cause mortality or disabling stroke at 2 years in patients who had attempted AVR. The trial randomly assigned 1,746 patients at 87 international centers. Bayesian statistical analysis was used, and the 24-month estimated endpoint for TAVR vs. surgery was 12.6% vs. 14% for robust noninferiority.
Of note, in this trial the surgical 30-day mortality was 1.6%, for an observed-to-expected ratio of 0.35, which is the lowest mortality from any of the randomized trials (see Table). Despite this exceptionally low surgical mortality, the 2-year mortality for TAVR vs. surgery was 11.2% vs. 11.5%, showing an excellent outcome for the TAVR arm. In addition to these longer-term outcomes, the TAVR arm had at 30 days significantly less stroke, transfusion, acute kidney injury and atrial fibrillation. The TAVR group had significantly less time in ICU, shorter hospital length of stay and better improvement in quality of life at 1 month.
These findings resulted in FDA approval of the self-expanding Evolut valve (Medtronic) as the second valve commercially available for the intermediate-risk population and should result in a class I recommendation in the next iteration of the valve guidelines, now that we have two randomized trials showing noninferiority.
Risk Level Crucial
Following the progression of the risk level in these successive trials is helpful for clinicians to understand what we should likely expect as we continue down the risk spectrum. These randomized trials have enrolled successively lower-risk populations. Once a patient has survived his or her initial procedure, longer-term survival is based on how well the aortic stenosis is treated and the initial risk level of the patient. Both TAVR and surgery treat aortic stenosis well, so survival after recovery from the procedure is based mainly on the patient’s intrinsic risk level. Looking at 1- and 2-year survival is therefore the best measure of the population risk level.
As seen in the Table, as a study population’s Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM) decreased, so did its 2-year mortality rate in both the surgery and TAVR groups. This impressive decline in 2-year mortality reflects the ever-decreasing initial population risk level. The question we ask is, what should we likely expect from the low-risk trials?
The CoreValve High-Risk trial was the first — and so far only — trial randomized against surgery that showed TAVR superiority for the powered primary endpoint. This superiority was unexpected and somewhat surprising.
This led us to look carefully at the timing and cause of death in this trial. For this column, we created an instantaneous risk for death plot and divided this into three sections: the procedural risk (0 to 30 days), the recovery risk (1 to 4 months) and longer-term survival after recovery (4 months to 1 year; see Figure). The rates parallel each other at all times except during the recovery period of 1 to 4 months, which was the only period for which survival was superior for TAVR. This appeared to be related to the inability of these high-risk patients to recover from the physiologic insult of surgery. A similar instantaneous HR for death for the SURTAVI intermediate-risk group shows that the rates parallel one another without the surgical rise in the recovery phase, presumably because these low-risk patients recover better from surgery. In the low-risk trials, we are likely to see the rates decrease in parallel and noninferiority seems likely.
Continued Investigation
Despite our thoughts about the potential outcomes of the low-risk trials, it is our responsibility as a field to complete these trials for a number of reasons we have previously enumerated. The results will help complete a journey — a relentless drive toward low risk — that has resulted in the best data available in the area of structural heart disease and arguably in cardiac interventions. We encourage those with an ability to enroll in the low-risk trials to do so, or to refer to centers completing this important work.
- References:
- Adams DH, et al. N Engl J Med. 2014;doi:10.1056/NEJMoa1400590.
- Cribier A, et al. Circulation. 2002;doi:10.1161/01.CIR.000047200.36165.B8.
- Gaudiani V, et al. J Thorac Cardiovasc Surg. 2017;doi:10.1016/j.jtcvs.2016.11.069.
- Leon MB, et al. N Engl J Med. 2016;doi:10.1056/NEJMoa1514616.
- Mumtaz M, et al. J Thorac Cardiovasc Surg. 2017;doi:10.1016/j.jtcvs.2017.02.035.
- Nishimura RA, et al. J Am Coll Cardiol. 2017;doi:10.1016/j.jacc.2017.03.011.
- Reardon MJ, et al. N Engl J Med. 2017;doi:10.1056/NEJMoa1700456.
- Smith CR, et al. N Engl J Med. 2011;doi:10.1056/NEJMoa1103510.
- For more information:
- Colin M. Barker, MD, is an interventional cardiologist and assistant professor at Houston Methodist DeBakey Heart and Vascular Center.
- Michael J. Reardon, MD, is professor of cardiothoracic surgery and Allison Family Distinguished Chair of Cardiovascular Research at Houston Methodist DeBakey Heart and Vascular Center. The authors can be reached at cmbarker@houstonmethodist.org.
Disclosures: Barker reports consulting or serving on an advisory board for Boston Scientific and Medtronic. Reardon reports that his institution received consultant fees from Medtronic and serving as principal investigator for two trials sponsored by Boston Scientific and two trials sponsored by Medtronic.