Skip to main content

Transcatheter vs. Surgical Aortic Valve Replacement After 10 Years

A multicenter randomized trial of transcatheter compared to surgical aortic valve replacement in low-risk older adult patients with symptomatic severe aortic stenosis has shown that major clinical outcomes and prosthetic valve failure were not different at 10 years of follow-up.

By Michael H. Crawford, MD, Editor

SYNOPSIS: A multicenter randomized trial of transcatheter compared to surgical aortic valve replacement in low-risk older adult patients with symptomatic severe aortic stenosis has shown that major clinical outcomes and prosthetic valve failure were not different at 10 years of follow-up.

SOURCE: Thyregod HGH, Jørgensen TH, Ihlemann N, et al. Transcatheter or surgical aortic valve implantation: 10-year outcomes of the NOTION trial. Eur Heart J 2024; Feb 7. doi: 10.1093/eurheartj/ehae043. [Online ahead of print].

The long-term durability of transcatheter aortic valve replacement (TAVR) compared to surgical aortic valve replacement (SAVR) with a bioprosthetic valve is unclear. Thus, the multicenter Nordic Aortic Valve Intervention Trial (NOTION) is of interest. NOTION enrolled 280 patients aged 70 years or older (mean age 79 years, 47% women) with symptomatic severe aortic stenosis (AS) and a mean Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM) score of 3.0, indicating a low-risk cohort. Patients with other significant cardiovascular disease or major organ failures were excluded. Outcomes were adjudicated at each center. Patients randomized to TAVR received a self-expanding first- or second-generation CoreValve bioprosthesis, using a transfemoral access in almost all cases. SAVR patients had a full sternotomy and standard implantation of a porcine or bovine stented bioprosthesis without the use of annular enlargement techniques. The primary outcome was a composite of all-cause mortality, stroke, or myocardial infarction (MI) at 10 years. Other clinical and echocardiographic outcomes also were assessed, including bioprosthetic valve deterioration and failure.

Four patients died before receiving a procedure, three TAVR patients crossed over to SAVR, and three SAVR patients did not receive a bioprosthesis, leaving a final population of 274 patients (139 TAVR and 135 SAVR). Ninety-nine percent of patients could be followed up at 10 years; of these, 36% were alive. Echocardiogram data were available in 81% of the patients alive at 10 years. The primary composite outcome was identical between the TAVR and SAVR groups (66%). Also, there was no significant difference in each component of the primary composite outcome.

Severe structural valve deterioration occurred in 1.5% of the TAVR patients and in 10% of the SAVR patients (P = 0.02). The cumulative incidence of severe bioprosthetic valve dysfunction was 21% and 43% (P < 0.001) after TAVR and SAVR, respectively, driven mainly by patient prosthesis mismatch, which was more common with SAVR (32% vs. 10% with TAVR, P < 0.001). The incidence of infective endocarditis was identical (7%), and no patients had clinical valve thrombosis. Bioprosthetic valve failure was not significantly different (10% TAVR and 14% SAVR, P = 0.4) and few patients underwent reintervention (4% TAVR vs. 2% SAVR). The authors concluded that in older adult patients with severe symptomatic AS and low surgical risk randomized to TAVR vs. SAVR, major clinical outcomes and valve failure were not different at 10 years of follow-up.


In the United States and Europe, more younger patients at low risk for surgery are undergoing TAVR despite guidelines that encourage surgery for low-risk younger patients because of uncertainty about the durability of TAVR.1,2 Thus, the 10-year results of NOTION are reassuring, since there was no difference in mortality, stroke, or MI between TAVR and SAVR. Early post-procedure, the incidence of new atrial fibrillation was higher with SAVR and pacemaker placement was higher with TAVR, both of which have been observed in other studies. Valve areas were larger, and gradients lower, after TAVR, but there were more paravalvular leaks compared to SAVR. However, valve failure and reintervention were infrequent and not different. Randomized trials of other newer TAVR valves in younger patients have only reported five-year data, and no differences in mortality compared to SAVR have been observed.

There were several limitations to NOTION. Frailty in these older patients was not formally assessed, and patients with significant coronary artery disease were excluded. The number of patients was small for a multicenter trial. The TAVR system used was an older self-expanding valve, which is believed to be inferior to newer systems. Also, the SAVR patients received one of five different bioprostheses that mainly employed porcine valve tissue, which many believe is inferior to bovine pericardium. Computed tomography scans for thrombus were not routinely performed, and echocardiogram data and stroke assessment were done at each site. In addition, more SAVR patients withdrew from the study, which may have introduced a bias. Finally, outcome determinations were not blinded.

The TAVR valves ranged in size from 26 mm to 31 mm and the SAVR valves ranged from 19 mm to 25 mm. Thus, it is not surprising that valve hemodynamics were better in the TAVR patients. However, these differences did not translate into better left ventricular ejection fractions (51% vs. 53%) or the regression of left ventricular hypertrophy. In fact, there are relatively few clinically significant differences between TAVR and SAVR in NOTION, and the major trial endpoints are not significantly different. Thus, at this time, it appears that TAVR is an acceptable procedure for older, low-risk surgical patients. However, more data are needed in younger patients for whom SAVR currently is the official recommendation, with mechanical valves preferred for those younger than 65 years of age.


  1. Vahanian A, Beyersdorf F, Praz F, et al. 2021 ESC/EACTS guidelines for the management of valvular heart disease. Eur Heart J 2022;43:561-632.
  2. Writing Committee Members; Otto CM, Nishimura RA, Bonow RO, et al. 2020 ACC/AHA guideline for the management of patients with valvular heart disease: A report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol 2021;77:e25-197.