Choice of intervention for severe calcific aortic stenosis

INTRODUCTION — For patients with severe calcific native aortic valve stenosis (AS) with an indication for valve replacement, intervention options include surgical aortic valve replacement (SAVR) or transcatheter aortic valve implantation (TAVI, also known as transcatheter aortic valve replacement or TAVR). A multidisciplinary team approach is recommended in approaching patients with severe AS, as the decisions involved are complex.

This topic will review the choice of therapy (SAVR, TAVI, or palliative medical therapy) for patients with severe AS with an indication for valve replacement.

Other related issues are discussed separately:

●Indications for valve replacement for severe AS. (See "Indications for valve replacement for high gradient aortic stenosis in adults".)

●Estimating the risk of aortic valve surgery. (See "Estimating the risk of valvular procedures".)

●Medical therapy of symptomatic AS. (See "Medical management of symptomatic aortic stenosis".)

●Percutaneous aortic valvuloplasty. (See "Percutaneous balloon aortic valvotomy for native aortic stenosis in adults".)

●Complications of TAVI. (See "Transcatheter aortic valve implantation: Complications".)

CHOICE OF INTERVENTION — For patients with severe calcific native AS with an indication for intervention, a choice is made between surgical aortic valve replacement (SAVR), transcatheter aortic valve implantation (TAVI), or palliative medical therapy based upon estimated surgical risk and other factors.

Initial assessment — Patients with severe calcific AS with an indication for valve replacement should undergo an assessment to determine whether the patient is likely to benefit from aortic valve intervention (SAVR or TAVI).

●The first step is to refer the patient for evaluation by a multidisciplinary heart valve team (including a cardiologist with expertise in structural valve interventions and a cardiothoracic surgeon).

●The heart valve team then assesses the patient's anticipated life expectancy with SAVR or TAVI and whether the patient's quality of life is likely to improve with SAVR or TAVI. If the patient has known potentially life-limiting disease other than AS (eg, cancer) [1,2], consultation with the clinicians involved in the care of this condition is important.

•If life expectancy with SAVR is >1 year and the patient's quality of life is likely to improve with SAVR or TAVI, the next step is evaluation by the Heart Valve Team of the risk of mortality and morbidity with SAVR (including the Society of Thoracic Surgeons Predicted Risk of Mortality [STS-PROM]) [3,4]. (See "Estimating the risk of valvular procedures".)

•If life expectancy with SAVR or TAVI is ≤1 year or the patient's quality of life is unlikely to improve with SAVR or TAVI, then palliative therapy with medical management is recommended. (See "Medical management of symptomatic aortic stenosis".)

Symptomatic patients — For patients with symptomatic severe calcific AS, our approach to choice of therapy (SAVR, TAVI, or no valve intervention) involves the following steps (algorithm 1):

●For patients with extreme surgical risk (≥50 percent probability of death or serious irreversible complication) or with an absolute contraindication to SAVR not captured by STS-PROM and in whom transfemoral TAVI is feasible, we recommend TAVI rather than medical therapy. For patients for whom transfemoral TAVI is not feasible, the Heart Valve Team should perform an individualized risk-benefit assessment of medical therapy versus alternative access TAVI. (See 'Contraindications to SAVR' below and 'TAVI versus medical treatment in inoperable patients' below.)

●For patients at high surgical risk (STS-PROM >8 with <50 percent probability of death), we determine whether transfemoral TAVI is feasible (see "Imaging for transcatheter aortic valve implantation", section on 'Determining eligibility for peripheral vascular access' and "Transcatheter aortic valve implantation: Periprocedural and postprocedural management", section on 'Access routes'):

•If yes, then we recommend transfemoral TAVI.

•If no, then the Heart Valve Team undertakes an individualized risk-benefit assessment of SAVR versus alternative (non-transfemoral) access TAVI. Factors to consider in the assessment are discussed below. (See 'Individualized risk-benefit assessment' below and "Transcatheter aortic valve implantation: Periprocedural and postprocedural management", section on 'Access routes'.)

●In patients with intermediate surgical risk (STS-PROM 4 to 8), we determine whether transfemoral TAVI is feasible and there is absence of high risk anatomic features (such as an adverse aortic root, low coronary ostia height, heavily calcified bicuspid aortic valve, and severe left ventricular outflow tract [LVOT] calcification) (see "Imaging for transcatheter aortic valve implantation" and "Transcatheter aortic valve implantation: Periprocedural and postprocedural management", section on 'Access routes'):

•If yes, we recommend transfemoral TAVI.

•If no, we recommend SAVR. If a relative contraindication to SAVR is present, an individualized risk-benefit assessment of SAVR versus alternative access TAVI is performed. (See 'Individualized risk-benefit assessment' below.)

●Optimum criteria for choice of intervention in patients with low surgical risk are uncertain. In patients with low surgical risk (ie, STS-PROM <4), we determine whether all four of the following criteria are met: age ≥65 years, transfemoral TAVI is feasible, aortic valve is trileaflet, and absence of high risk anatomic features (such as an adverse aortic root, low coronary ostia height, or severe LVOT calcification) (see 'In low-risk symptomatic patients' below).

•If yes, we recommend transfemoral TAVI.

•If no, SAVR is preferred. If a relative contraindication to SAVR is present (unusual in this setting), then an individualized risk-benefit assessment of SAVR versus TAVI is performed. (See 'Individualized risk-benefit assessment' below.)

The above approach is similar to that in the 2020 American College of Cardiology/American Heart Association (ACC/AHA) valve guideline and the 2021 European Society of Cardiology valve guidelines [5,6]. (See 'TAVI versus SAVR' below.)

Asymptomatic patients — For patients with asymptomatic severe AS with an indication for valve replacement (eg, presence of LV ejection fraction (LVEF) <50 percent, very severe AS, or fall in systemic blood pressure with exercise), the choice of SAVR versus TAVI is based upon an individualized risk-benefit assessment including the estimated surgical risk (with intermediate or higher estimated surgical risk favoring TAVI and low estimated surgical risk favoring SAVR) and estimated risk of complications with TAVI (with feasibility of transfemoral TAVI and absence of high risk anatomic features favoring TAVI). For asymptomatic patients with severe AS and an abnormal exercise test, very severe AS, rapid progression, or an elevated B-type natriuretic peptide, the 2020 ACC/AHA valve guidelines favor SAVR rather than TAVI [5]. There is limited evidence comparing SAVR versus TAVI in asymptomatic patients as these patients were excluded from most of the supporting trials. (See 'Individualized risk-benefit assessment' below and 'Outcomes' below.)

Asymptomatic patients with severe AS who lack an indication for valve replacement should have frequent clinical evaluation to monitor for symptom onset so that timely treatment is provided for symptomatic patients. (See "Medical management of asymptomatic aortic stenosis in adults", section on 'Serial evaluation'.)

PROCEDURAL RISK ASSESSMENT — Procedural risk assessment is performed after the initial assessment described above. (See 'Initial assessment' above.)

Estimation of surgical risk

General approach — Operative risk assessment (including identification of high and prohibitive risk) includes consideration of estimated operative mortality (eg, using the STS Short-Term Risk Calculator), frailty, compromised major organ systems, and comorbidities (table 1). Accurate estimation of the risk of surgical aortic valve replacement (SAVR) performed by an experienced cardiothoracic surgeon and multidisciplinary valve team is vital to appropriate evaluation of potential candidates. Risk assessment for valvular surgery is discussed further separately. (See "Estimating the risk of valvular procedures".)

Contraindications to SAVR — SAVR is not a treatment option in patients with extreme surgical risk (ie, ≥50 percent probability of death or serious irreversible complication) or with an absolute contraindication not captured by Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM).

Absolute contraindications not captured by STS-PROM include "porcelain aorta" (heavy circumferential calcification or severe atheromatous plaques of the entire ascending aorta extending to the arch such that aortic cross-clamping is not feasible) and hostile chest (chest conditions such as severe radiation damage or complications from prior surgery that make operation through sternotomy or thoracotomy prohibitively hazardous) [7].

Relative contraindications not captured by STS-PROM include frailty, severe liver disease/cirrhosis, prior coronary artery bypass graft surgery with vulnerable graft location as assessed by computed tomography, and severe pulmonary hypertension or severe right ventricular dysfunction [7].

Estimation of TAVI risk

General approach — Estimates of TAVI procedural risk are based upon data from registries and clinical trials. Individualized assessment comparing TAVI and SAVR is discussed below. (See 'Individualized risk-benefit assessment' below.)

Patients with increased risk of TAVI complications include those who are not candidates for transfemoral TAVI and those with relative contraindications to TAVI as described below. (See 'By access site' below and 'Contraindications to TAVI' below.)

Of note, TAVI outcomes may also be related to institutional factors, with an inverse relationship observed between procedural volume and 30-day mortality, as discussed separately. (See "Transcatheter aortic valve implantation: Periprocedural and postprocedural management", section on 'Preprocedural considerations'.)

The effect of bicuspid aortic valve (other than heavily calcified bicuspid valve) on TAVI outcomes is uncertain and is of less impact with the use of newer-generation transcatheter valves, as discussed separately (see "Bicuspid aortic valve: Intervention for valve disease or aortopathy in adults", section on 'Transcatheter aortic valve implantation' and "Bicuspid aortic valve: Intervention for valve disease or aortopathy in adults"). While a congenitally bicuspid aortic valve was previously considered an exclusion criterion for TAVI, TAVI has been successfully performed in many patients with this disorder, as discussed separately. Once severe calcific stenosis is present, reliable identification of the number of valve leaflets is problematic, so it is likely that TAVI has been performed in many patients with a congenital bicuspid valve.

Contraindications to TAVI — Patients with a number of conditions are generally excluded from transcatheter aortic valve replacement (TAVI) [8,9].

Absolute contraindications include the following clinical issues:

●Estimated life expectancy <12 months due to noncardiac comorbid conditions.

●Improvement of quality of life by TAVI unlikely because of comorbidities.

●Severe other valve disease with major contribution to the patient's symptoms if treatable only by surgery.

●Additional absolute contraindications are related to anatomic issues and clinical conditions:

•Inadequate annulus size: Native aortic annulus size as measured by multidetector computed tomography (MDCT) is <18 mm (for a native valve), <17 mm (for a surgical valve), or >the largest annulus size for which a TAVI device is available (30 mm). This criterion is subject to change as the range of available device sizes changes. In clinical practice, inability to select an appropriate valve due to size issues is a very rare event.

Note that valve size numbers do not correspond to actual annular measurements. Deciding which size valve to select for a given manufacturer is complex and is made using MDCT imaging area and perimeter measurements, with different metrics preferred for different valve types.

•Active endocarditis.

•Elevated risk of coronary ostium obstruction which is associated with presence one or more of the following anatomic features: asymmetric valve calcification, low coronary ostia height, adverse aortic root (small sinus of Valsalva or small, calcified sinotubular junction). (See "Transcatheter aortic valve implantation: Complications", section on 'Coronary obstruction' and "Imaging for transcatheter aortic valve implantation", section on 'Coronary ostia'.)

Relative contraindications include the following:

●Evidence (such as creatine kinase [CK] plus CK-MB elevation and/or troponin elevation) of an acute myocardial infarction within one month before the intended treatment.

●Untreated coronary artery disease requiring revascularization, particularly if coronary artery bypass graft surgery is deemed necessary.

●Hemodynamic instability requiring inotropic support, or mechanical heart assistance within 30 days of screening evaluation.

●Respiratory instability requiring mechanical ventilation within 30 days of screening evaluation.

●Need for emergency surgery.

●Hypertrophic cardiomyopathy with or without obstruction.

●Severe pulmonary hypertension and right ventricular dysfunction.

●A known contraindication or hypersensitivity to all anticoagulation regimens or inability to be anticoagulated for the study procedure.

●Echocardiographic evidence of vegetation (unless completely healed).

●Elevated risk of annular rupture which is associated with presence of one or more of the following features: bulky calcification of either aortic annulus or leaflets, heavily calcified bicuspid valve, heavily calcified LV outflow tract (LVOT), or severe asymmetric subaortic LV hypertrophy. An adverse aortic root is a risk factor for annular rupture as well as coronary artery obstruction. (See "Transcatheter aortic valve implantation: Complications", section on 'Annular rupture'.)

●Contraindications to particular access sites are discussed separately. (See "Transcatheter aortic valve implantation: Periprocedural and postprocedural management", section on 'Access routes'.)

Individualized risk-benefit assessment — Decision-making by a multidisciplinary heart valve team should take into consideration a patient's life expectancy, frailty, comorbidities, specific anatomy, values, and preferences. An individualized risk-benefit assessment of SAVR versus TAVI is performed in the following clinical settings:

●For symptomatic patients in the following settings (see 'Symptomatic patients' above):

•Patients with high surgical risk who are not candidates for transfemoral TAVI, an individualized risk-benefit analysis is performed comparing SAVR and alternative access TAVI.

•Patients with intermediate or low surgical risk, are not candidates for transfemoral TAVI or have a high-risk anatomic feature (for low risk patients, this includes presence of a unicuspid or bicuspid valve), and who have a relative contraindication for SAVR, an individualized risk-benefit analysis is performed comparing transfemoral TAVI versus SAVR.

●For asymptomatic patients with AS who have an indication for valve replacement. (See 'Asymptomatic patients' above.)

Clinical factors influencing the choice between TAVI and SAVR include the patient's values and preferences, age, estimated life expectancy with AVR, as well as presence of concurrent conditions. An issue influencing the choice of TAVI versus SAVR is lack of data on very long-term outcomes with TAVI.

●The following factors favor SAVR (see "Choice of prosthetic heart valve for surgical aortic or mitral valve replacement", section on 'Recommendations for valve choice'):

•Another indication for cardiac surgery (eg, need for coronary artery bypass graft or mitral valve surgery).

•Patient age <75 years old.

•Patient characteristics favoring mechanical valve replacement (eg, age <55 with long life-expectancy plus absence of contraindications for anticoagulation). Mechanical valves can be implanted only surgically.

•An anatomic feature that would make TAVI high risk (such as adverse aortic root [small sinus of Valsalva or small, calcified sinotubular junction], low coronary ostia height, and severe LVOT calcification or severely calcified bicuspid valve).

●The following factors favor TAVI:

•Patient age ≥75 years.

•Transfemoral TAVI is feasible.

•Risk factors for SAVR, including those not captured by the STS-PROM score (such as frailty, prior mediastinal irradiation, or cirrhosis).

•Female sex – Although females experience more major bleeding and vascular complications with TAVI, female sex is an independent predictor of lower mortality after TAVI compared with females undergoing surgery (as well as compared with males undergoing TAVI). Thus, sex-specific mortality risk following TAVI is opposite of that following SAVR, for which females have higher mortality risk than males. We suggest including female sex as a factor since a mortality benefit has been observed with TAVI versus SAVR among females with intermediate to high surgical risk. (See 'Sex-specific differences' below.)

SELECTION OF TAVI VALVE TYPE

Patient-specific considerations — For the majority of patients undergoing transcatheter aortic valve implantation (TAVI), any one of the commercially available valves is suitable. In the United States, these include the balloon expandable (SAPIEN 3 and SAPIEN 3 Ultra) valves and the self-expanding valves (Evolut R, Evolut PRO, and PRO+). Approved valves in Europe include the preceding valves as well as the balloon-expandable Myval valve, the self-expanding ACURATE neo and neo 2, and Portico valves (the latter is repositionable prior to deployment).

For patients treated at a center having sufficient experience with and access to many types of valves, there are certain patient-specific issues that might influence the choice of valve system type:

●Most valve types, but not all, cover the full range of annulus size.

●In a patient deemed to be at high risk of annulus rupture (eg, a patient with a small, highly calcified annulus), a self-expanding rather than a balloon-expandable valve may be chosen to reduce the risk of annular rupture (as one of several potential strategies to attempt to reduce the risk of rupture). Annular rupture has been observed almost exclusively after use of a balloon-expandable valve and very rarely after use of a self-expandable valve [10]. (See "Transcatheter aortic valve implantation: Complications", section on 'Annular rupture'.)

●If there are concerns about coronary obstruction, then a valve system with recapturable technology may be favored.

●When performing a valve-in-valve procedure to treat a small surgical bioprosthetic valve, a supra-annular TAVI valve might offer greater effective orifice area.

●If there is a spur of asymmetric calcification protruding into the outflow tract, then the choice of a valve with external sealing skirt may be preferable.

Center-specific considerations — Operators have generally selected the type of TAVI valve to implant based upon local practice, operator training, medical center experience, and availability (based upon the regulatory approval status) rather than specific patient-related factors. Individual center procedure volume is an important factor in establishing and maintaining optimum patient outcomes. As a consequence, many centers have acquired experience with three or less valve systems. While this situation is changing, particularly in countries where there are multiple approved devices, maintenance of sufficient experience with each device used continues to be important for optimum patient outcomes. The differences in patient selection and procedural steps among competing device types are greater for various TAVI systems than for most other interventional cardiovascular procedures.

OUTCOMES — Transcatheter aortic valve implantation (TAVI) indications and techniques are rapidly evolving with published randomized trials in patients with symptomatic AS in patients with extreme, high, intermediate, and low surgical risk as described below.

Evidence on the use of TAVI in asymptomatic patients with AS is largely indirect, as these patients were excluded from randomized trials in intermediate and higher surgical risk patients and only a small minority of patients in the low surgical risk trials were asymptomatic. The ongoing EARLY TAVR trial is comparing TAVI with clinical surveillance in patients with asymptomatic severe AS [11].

A systematic review found very limited evidence on the values and preferences of adults with AS [3]. The available evidence suggested considerable variability in individual values and preferences, which highlights the importance of shared decision-making.

TAVI versus medical treatment in inoperable patients — The Placement of Aortic Transcatheter Valves (PARTNER) multicenter trial (cohort B) in 358 inoperable patients with severe AS demonstrated a significant reduction in mortality and improvement in functional class in inoperable patients with balloon-expandable TAVI compared with standard medical care [12-15]. Enrolled patients were randomly assigned to either standard therapy (including balloon aortic valvotomy) (see "Percutaneous balloon aortic valvotomy for native aortic stenosis in adults") or TAVI with an Edwards SAPIEN valve via transfemoral approach. The mean age was 83 years and the mean Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM) was 11.7 percent. (See "Estimating the risk of valvular procedures".)

The following findings were noted:

●At one year, the mortality rate was reduced with TAVI compared with standard therapy (30.7 versus 50.7 percent). At two years, the mortality rates for TAVI and standard therapy were 43.4 and 68 percent; at three years, the mortality rates were 54.1 percent and 80.9 percent, and at five years, the mortality rates were 71.8 and 93.6 percent.

●Among survivors at one, two, three, and five years, functional class was better with TAVI versus standard therapy (eg, 86 versus 60 percent in New York Heart Association (NYHA) functional class I or II at five years).

●The stroke rate was significantly higher in the TAVI group than in the standard therapy group at 30 days (6.7 versus 1.7 percent), at two years (13.8 versus 5.5 percent), and three years (15.7 versus 5.5 percent). At five years, the risk of stroke was similar in the TAVI and standard therapy groups (16 versus 18.2 percent). A competing risk analysis showed no continuous hazard of stroke associated with TAVI after the initial procedural risk [15].

●In the standard therapy group, balloon aortic valvotomy was performed in 82.3 percent at one year and in 85.3 percent at two years. Moderate or severe transvalvular aortic regurgitation was observed in 16.9 percent at 30 days and 15.2 percent at one year.

●In the TAVI group, moderate or severe paravalvular aortic regurgitation was identified in 12.4 percent at 30 days, in 8.8 percent at one year, and in 4.5 percent at three years.

Given the results of the PARTNER cohort B study, it was determined that a randomized trial comparing self-expanding TAVI and medical therapy could not be performed. The CoreValve Extreme Risk United States Pivotal Trial, a prospective single-arm study, compared TAVI with the self-expanding CoreValve to a pre-specified estimate of 12-month mortality or major stroke with medical therapy (43 percent, based upon results of a meta-analysis and data from the PARTNER cohort B) [16].

●For the 489 patients (mean age 83 years; mean STS-PROM 10.3) who underwent attempted transfemoral CoreValve transcatheter heart valve implantation, the rate of all-cause mortality or major stroke at 12 months was 26 percent, which was significantly lower than the prespecified performance goal of 43 percent.

●Procedural events at 30 days included life-threatening/disabling bleeding in 12.7 percent, major vascular complications in 8.2 percent, and need for permanent pacemaker placement in 21.6 percent.

TAVI versus SAVR — While individual trials comparing TAVI versus SAVR have targeted specific surgical risk groups, meta-analyses have pooled these groups to examine pooled outcomes and subgroup differences. A meta-analysis included four trials (the largest trial with 2032 intermediate-risk patients, two studies with 699 and 795 high-risk patients, and one study with 280 low-risk patients) to assess outcomes at two years [17].

●Mortality was reduced with TAVI as compared with SAVR (hazard ratio [HR] 0.87, 95% CI 0.76-0.99) with homogeneity across the trials. The mortality reduction with TAVI versus SAVR was robust among patients undergoing transfemoral TAVI (HR 0.80, 95% CI 0.69-0.93) but not present among patients undergoing transthoracic TAVI. Similarly, there was a borderline significant interaction for sex (p = 0.05), with a significant mortality reduction with TAVI versus SAVR among females (HR 0.68, 95% CI 0.50-0.91) but not among males (HR 0.99, 95% CI 0.77-1.28).

●Acute kidney injury, new-onset atrial fibrillation, and major bleeding were less frequent with TAVI while major vascular complications, incidence of permanent pacemaker implantation, and paravalvular regurgitation were less frequent with SAVR.

With high surgical risk — In the above described meta-analysis in which nearly all patients had intermediate to high surgical risk, transfemoral TAVI (but not transthoracic TAVI) was associated with lower mortality than SAVR [17]. The meta-analysis included the following two pivotal trials in patients with high surgical risk:

●In cohort A of the PARTNER randomized trial, 699 high-risk patients were randomly assigned to balloon-expandable TAVI or SAVR [18-20]. The mean age was 84 years and the mean STS-PROM was 11.7 percent. There were similar mortality rates in the two treatment groups at 30 days (3.4 and 6.5 percent, p = 0.07), one year (24.3 and 26.8 percent), two years (33.9 and 35 percent), and five years (67.8 and 62.4 percent), but higher rates of moderate to severe aortic regurgitation following TAVI at 30 days, one year, and two years (6.9 versus 0.9 percent at two years; 14 versus 1 percent at five years). The presence of paravalvular aortic regurgitation was associated with increased late mortality, as discussed separately. (See "Transcatheter aortic valve implantation: Complications", section on 'Paravalvular regurgitation'.)

●The United States CoreValve High Risk randomized trial compared self-expanding TAVI and SAVR in 795 high-risk patients with AS. The mean age was 83.2 years and the mean STS-PROM was 7.4 percent. This trial reported a significantly lower mortality rate with TAVI (14.2 versus 19.1 percent at one year; 22.2 versus 28.6 at two years) [21-23]. The rate of death or major stroke was significantly lower with TAVI at one (16.3 versus 22.5 percent), two (24.2 versus 32.5 percent), and three years (35.0 versus 41.6 percent).

In both trials, major vascular complications were more frequent after TAVI, and major bleeding and new-onset atrial fibrillation were more frequent after SAVR.

Since data are not convincing that outcomes are different with different types of transcatheter heart valves, TAVI (with either balloon-expandable or self-expanding valve) is recommended for high-risk patients who are candidates for a transfemoral approach. (See 'Symptomatic patients' above.)

In intermediate-risk patients — Randomized trials in patients with severe AS with intermediate surgical risk have found similar rates of death or disabling stroke following TAVI and SAVR [24,25] as summarized by the following meta-analyses:

●A meta-analysis comparing TAVI and SAVR in patients with severe AS with predominantly intermediate surgical risk found similar mortality rates at 30 days (3.0 versus 3.0 percent; relative risk [RR] 1.03; 95% CI 0.71-1.48), one year (9.6 versus 9.6 percent; RR 1.01; 95% CI 0.08-1.28) and ≥2 years (14.2 versus 13.5 percent; RR 1.01; 95% CI 0.62-1.66) in the two treatment groups (data shown based upon analysis in randomized trials) [26]. In analyses that included observational studies as well as randomized trials, the point estimate for stroke suggested a benefit for TAVI compared with SAVR but the confidence interval overlapped no effect at 30 days, one year, and ≥2 years (total stroke events 5.7 versus 6.4 percent; RR 0.92; 95% CI 0.80-1.05). This meta-analysis did not examine the relationship between access site and outcomes.

●A separate meta-analysis of four randomized trials with 3179 patients with severe AS at predominantly intermediate risk of perioperative death examined the effect of TAVI versus SAVR on outcomes at median follow-up of two years [27]. The included trials were the PARTNER 2A, US Pivotal, NOTION, and STACCATO trials [21,24,28,29]. Baseline risk estimates were derived from a systematic review of observational studies of bioprosthetic SAVR [30].

•Transfemoral TAVI compared with SAVR resulted in reduced mortality (HR 0.79, 95% CI 0.88-0.94; 30 fewer per 1000 patients) and reduced acute kidney injury (RR 0.38, 95% CI 0.27-0.53, 53 fewer per 1000 patients). The point estimate for stroke also suggested a benefit for transfemoral TAVI compared with SAVR but the confidence interval overlapped no effect (RR 0.80, 95% CI 0.61-1.01, 20 fewer per 1000 patients).

•TAVI compared with SAVR resulted in reduced atrial fibrillation and major bleeding, with greater reductions in bleeding among patients undergoing transfemoral TAVI versus transapical TAVI.

•In contrast, TAVI compared with SAVR resulted in more frequent worsened symptoms of heart failure (one point worse on the NYHA scale; odds ratio [OR] 1.29, 95% CI 1.08-1.55, 59 more per 1000 patients), aortic valve re-intervention (RR 3.25; 95% CI 1.29-8.14, 7 more per 1000 patients), permanent pacemaker insertion (RR 2.45, 95% CI 1.17-5.14, 134 more per 1000 patients), and moderate or severe aortic valve regurgitation (RR 12.22, 95% CI 5.17-28.88, 80 more per 1000 patients).

•For alternative (nontransfemoral) access TAVI compared with SAVR, the point estimates suggested increased mortality (HR 1.34, 95% CI 0.91-1.97, 57 more per 1000 patients) and stroke (HR 1.67, 95% CI 0.97-2.87, 45 more per 1000) but the confidence intervals overlapped with no effect.

Limitations of studies included in these meta-analyses include use of transcatheter valves that are no longer in use (and cause higher rates of paravalvular aortic regurgitation than current models), use of transthoracic access routes that are now used less commonly than the subclavian/axillary approach, use of only bioprosthetic valves for SAVR in the included randomized trials, and limited duration of follow-up. In addition, in the second meta-analysis, the transthoracic component of the meta-analysis pooled data from the STACCATO trial (comparing transapical TAVI with SAVR) with the transthoracic subgroup of the PARTNER 2A trial. Limitations of the STACCATO trial include use of transapical access, which is a less common route, lack of preprocedural multidetector computed tomography (unlike PARTNER 2A and current clinical practice), the device success rate was unusually low (79 percent) compared with rates seen in higher-risk patients, and the rate of adverse events was unusually high, which triggered early termination of the study [29].

The largest trial included in both meta-analyses is the PARTNER 2A trial, which randomly assigned 2032 intermediate-risk patients (mean STS-PROM score of 5.8; mean age 82 years) with severe AS to undergo either TAVI (with the SAPIEN-XT balloon-expandable valve, which is no longer implanted) or SAVR with two-year follow-up [24]; a study published after these meta-analyses reported five-year outcomes [31]. Patients were divided into two cohorts prior to randomization on the basis of an evaluation of the peripheral arteries: 76.3 percent were included in the transfemoral-access cohort and 23.7 percent were included in the transthoracic-access (transapical or transaortic) cohort.

●Early outcomes [24]:

•Early benefits of TAVI included lower 30-day rates of acute kidney injury (1.3 versus 3.1 percent), severe bleeding (10.4 versus 43.4 percent), and new onset atrial fibrillation (9.1 versus 26.4 percent).

•Early benefits of SAVR included lower rates of major vascular complications (5 versus 7.9 percent at 30 days) and less paravalvular aortic regurgitation (moderate or severe in 0.6 versus 3.7 percent at 30 days). Patients in the TAVI group with moderate or severe paravalvular aortic regurgitation at 30 days had higher mortality during two- and five-year follow-up than did patients with no or trace aortic regurgitation [24,31].

●Late outcomes:

•The rate of death from any cause or disabling stroke was similar in the TAVI and SAVR groups. The Kaplan-Meier event rates in the two groups were similar at two years (19.3 and 21.1 percent; HR 0.89; 95% CI 0.73-1.09) [24] and five years (47.9 and 43.4; HR 1.09; 95% CI 0.95-1.25), but there were more events between two and five years in the TAVI group (HR 1.27; 95% CI 1.06-1.53) [31]. Differing results were seen in the two access cohorts:

-In the transfemoral-access cohort, TAVI resulted in a lower event rate than SAVR at two years (16.8 and 20.4 percent; HR 0.79; 95% CI 0.62-1.00); however, at five years, event rates were similar (44.5 versus 42.0 percent; HR 1.02; 95% CI 0.87-1.20).

-In the transthoracic-access cohort, outcomes were similar in the TAVI and SAVR groups at two years (27.7 and 23.4 percent; HR 1.21; 95% CI 0.79-1.65); in contrast, at five years, events rates were higher with TAVI (59.3 versus 48.3 percent; HR 1.32; 95% CI 1.02-1.71).

•Improvement in health status at five years was similar in the TAVI and SAVR groups (NYHA functional class I or II in 89 and 92.7 percent).

•A late benefit of SAVR compared with TAVI was lower rates of repeat hospitalizations (25.2 versus 33.3 percent) at five years [31] and lower rates of at least mild paravalvular AR (6.3 versus 33.3 percent).

•Rates of aortic valve reinterventions at five years were lower after SAVR (0.8 versus 3.2 percent) [31]. However, reinterventions after TAVI were due to progressive stenosis or regurgitation, and nearly all (18 of 21) were treated with either TAVI or balloon valvuloplasty; in-hospital mortality from valve reintervention was 5 percent (1 of 21 patients). In contrast, reinterventions after SAVR were largely due to endocarditis (four of six cases), and most were treated with repeat surgery; in-hospital mortality was 50 percent (three of six patients).

Of note, these data are not sufficient to compare the long-term risk of endocarditis after TAVI and SAVR. The risk of endocarditis after TAVI and SAVR is discussed separately. (See "Prosthetic valve endocarditis: Epidemiology, clinical manifestations, and diagnosis" and "Transcatheter aortic valve implantation: Complications".)

Additional data on outcomes in intermediate surgical risk patients were provided by the SURTAVI trial (included in the first meta-analysis described above but not the second one), which randomly assigned 1746 patients with symptomatic severe AS with intermediate surgical risk (mean STS-PROM 4.5 percent) to TAVI (with a self-expanding bioprosthesis: CoreValve in 84 percent and Evolut R in 16 percent) or SAVR; 1660 patients underwent attempted SAVR or TAVI [25]. Nearly all of the TAVI procedures (93.6 percent) were performed via iliofemoral access.

●The incidence of the primary composite end point of death from any cause or disabling stroke at 24 months was similar in the TAVI and SAVR groups (12.6 and 14.0 percent).

●NYHA symptoms and quality of life (measured by the Kansas City Cardiomyopathy Questionnaire [KCCQ] summary score) improved significantly in both groups through 24 months of follow-up. At one month, there was a higher proportion of patients with an improved KCCQ summary score at one month in the TAVI group compared with the SAVR group.

●The SAVR treatment group experienced significantly higher 30-day rates of acute kidney injury (4.4 versus 1.7 percent), atrial fibrillation (43.4 versus 12.9 percent), and transfusion requirement (41.1 versus 12.6). The TAVI group experienced significantly higher 30-day rates of major vascular complications (6.0 versus 1.1 percent) and need for permanent pacemaker implantation (25.9 versus 6.6 percent). Moderate or severe paravalvular aortic regurgitation was more common at one year in the TAVI group (5.3 versus 0.6 percent in the SAVR group). Mean prosthetic valve gradients were significantly lower and prosthetic aortic valve areas were higher in the TAVI group.

An observational study using a propensity score analysis suggested that TAVI with a balloon-expandable SAPIEN XT valve may be superior to SAVR for intermediate-risk patients, but it is possible that residual confounders influenced the results [32].

In low-risk symptomatic patients — The efficacy and safety of TAVI in patients with AS with low estimated surgical risk were evaluated by the following randomized trials [33,34]:

●Self-expanding valve – In the Evolut Low Risk trial, 1468 patients (mean age 74) with severe AS and low surgical risk (mean STS-PROM 1.9±0.7) were randomly assigned to TAVI with a self-expanding valve or SAVR; a total of 1403 patients underwent the assigned procedure [33].

Nearly all TAVI procedures were performed via transfemoral access (99 percent). Percutaneous coronary intervention (concomitant or staged) was performed in 6.9 percent of the patients undergoing TAVI. Among patients undergoing SAVR, 26.3 percent underwent a concomitant surgical procedure (most commonly coronary artery bypass grafting, which was performed in 13.6 percent).

•At 30 days, mortality rates were similar with TAVI and SAVR (0.5 versus 1.3 percent). The TAVI group had significantly lower incidences of disabling stroke (0.5 versus 1.7 percent), bleeding complications (2.4 versus 7.5 percent), acute kidney injury (0.9 versus 2.8 percent), and atrial fibrillation (7.7 versus 35.4 percent) but higher rates of moderate or severe aortic regurgitation (3.5 versus 0.5 percent) and permanent pacemaker implantation (17.4 versus 6.1 percent).

•At one year, mortality rates were similar in the TAVI and SAVR groups (2.4 versus 3.0 percent). Hospitalizations for HF were less frequent in the TAVI group (3.2 versus 6.5 percent) and prosthetic aortic valve gradients were lower with TAVI (8.6 mmHg versus 11.2 mmHg) than in the surgery group.

The estimated incidence of the primary endpoint (a composite of death or disabling stroke at two years) was 5.3 percent in the TAVI group and 6.7 percent in the surgery group (absolute difference 1.4 percentage points; 95% credible interval for difference -4.9 to 2.1). Thus, the noninferiority threshold was met.

•At four years, rates of mortality (9.0 versus 12.1 percent), disabling stroke (2.9 versus 3.8 percent), and aortic valve rehospitalization (10.3 versus 12.1 percent) were similar in the TAVI and SAVR groups [35]. However, the composite of these outcomes was lower in the TAVI group (18.0 versus 22.4 percent; HR 0.78, 95% CI 0.61-0.98). Mean aortic valve gradients continued to be lower after TAVI (9.8 versus 12.1 mmHg). New permanent pacemaker implantation was more frequent in the TAVI group (24.6 versus 9.9 percent). One limitation was the higher rate of follow-up after TAVI than after SAVR (94.7 versus 89.2 percent).

●Balloon-expandable valve – In the PARTNER 3 trial, 1000 patients (mean age 73) with severe AS and low surgical risk (mean STS-PROM 1.9±0.7) were randomly assigned to TAVI with a balloon-expandable valve or SAVR; 950 patients received the assigned procedure [34].

All TAVI procedures were performed via transfemoral access. Among patients undergoing TAVI, a concomitant procedure was performed in 7.9 percent (most commonly percutaneous coronary intervention, which was performed in 6.5 percent). Among patients undergoing SAVR, 26.4 percent underwent a concomitant procedure (most commonly coronary artery bypass grafting, which was performed in 12.8 percent).

•At 30 days, mortality rates with TAVI and SAVR were similar (0.4 versus 1.1 percent). TAVI resulted in lower rates of stroke (0.6 versus 2.4 percent) and new-onset atrial fibrillation (5 versus 39.5 percent), and there were no significant differences in the frequency of permanent pacemaker insertions (6.6 versus 4.1 percent) or moderate or severe paravalvular regurgitation (0.8 versus 0.0).

•At one year, mortality rates (1.0 versus 2.5 percent), prosthetic valve mean gradients (13.7 versus 11.6), and frequency of moderate or severe paravalvular regurgitation (0.6 versus 0.5 percent) were similar in the TAVI and surgery groups. The estimated incidence of the primary endpoint (a composite of death, stroke, or rehospitalization at one year) was significantly lower in the TAVI group than in the surgical group (8.5 versus 15.1 percent; absolute difference -6.6 percentage points, 95% CI -10.8 to -2.5). Thus, both noninferiority and superiority criteria were met.

•At two years, mortality rates (2.4 versus 3.2 percent) and stroke rates (2.4 versus 3.6 percent) were similar in the two groups [36]. The risk of valve thrombosis was higher after TAVI than after surgery (2.6 versus 0.7 percent). The primary endpoint remained lower in the TAVI group than in the surgical group (11.5 versus 17.4 percent).

•At five years, rates of mortality (10.0 versus 8.2 percent), stroke (5.8 versus 6.4 percent), and rehospitalization (13.7 versus 17.4 percent) were similar in the two groups [37]. The primary endpoint was nominally, but not significantly, lower in the TAVI group than in the surgical group (22.8 versus 27.2 percent). Bioprosthetic valve failure also occurred at similar rates in the TAVI and surgical groups (3.3 versus 3.8 percent). Valve thrombosis was rare and was more common in the TAVI group (2.5 versus 0.2 percent).

These trials indicate generally favorable results of TAVI to four and five years for selected low surgical risk patients with anatomical suitability for TAVI [35,37]. Additional data, including longer term outcomes, will further inform the choice of intervention for patients with low surgical risk [38].

Differences between the two trials limit the validity of trial comparisons. There were some differences in patient populations as well as composite endpoints between the Evolut Low Risk (self-expanding valve) and PARTNER 3 (balloon-expandable valve) trials. In the PARTNER 3 trial, a greater proportion of screened patients were excluded for risk factors such as severe left ventricular outflow tract calcium, adverse aortic root (small sinus of Valsalva or small, calcified sinotubular junction), and poor transfemoral access; these exclusions may have contributed to the low rates of TAVI complications, such as need for permanent pacemaker insertion and paravalvular regurgitation, in this trial. The PARTNER 3 trial included rehospitalization at one year in the composite endpoint but the Evolut Low Risk trial did not. However, the Evolut Low Risk trial reported a reduction in hospitalizations for HF at one year, so one-year outcomes for TAVI with the self-expanding valve and the balloon-expandable valve may be similar. (See 'Transcatheter valve type' below.)

Data from registries — Additional information on outcomes following TAVI come from registry studies. Reports from the Society of Thoracic Surgeons/American College of Cardiology (STS/ACC) registry, the United Kingdom Transcatheter Aortic Valve Implantation (UK TAVI) registry, and the German Aortic Valve Registry (GARY) have included the following outcome data for patients with intermediate to high median risk (ie, STS-PROM 7.1 [39], 5 [40], or logistic Euroscore 18.5 [41,42]):

●Early mortality rates of 5.2 percent in-hospital [40] and 7 percent at 30 days [39,41].

●In-hospital stroke rates were 1.9 and 4.1 percent in two of these studies [39,41].

●One-year mortality rates were 23.7 and 21.4 percent in two of these studies [39,42].

●In the United Kingdom (UK) registry, mortality at two, three, five, and six years was 26.3, 38.8, 54.5, and 62.7 percent [41-43].

In the UK TAVI registry, stroke within 30 days of TAVI was the only independent procedural predictor of mortality at three and five years [41]. Independent predictors of three-year mortality were renal dysfunction, atrial fibrillation, respiratory dysfunction, and ventricular dysfunction. Coronary artery disease and age were independent predictors of mortality at five years. Device type, access route, and paravalvular leak did not independently predict long-term outcome.

A study of 241 patients (mean age 79 years) from the UK TAVI registry found that 91 percent of patients were free of structural valve degeneration at 5 to 10 (median 5.8)-year echocardiographic follow-up [44]. There was one case of new severe aortic regurgitation at 5.3 years, 12 cases of moderate aortic regurgitation, and nine cases of moderate restenosis.

Subgroup differences

By access site — As described above, patients undergoing transfemoral TAVI have better outcomes than patients undergoing alternative (nontransfemoral) access TAVI, as indicated by subgroup analyses of meta-analyses and individual trials. As an example, a meta-analysis of 27 observational studies and one randomized trial with a total of 17,020 patients undergoing TAVI found that 30-day mortality was 4.7 percent with the transfemoral approach and 8.1 percent with an alternative approach [45]. One-year mortality was 16.4 percent with transfemoral access and 24.8 percent with nontransfemoral access. Transfemoral access was associated with a higher rate of vascular complications (OR 2.1; 95% CI 1.48-2.99) but a lower rate of surgical conversion (OR 0.59; 95% CI 0.42-0.81), while rates of bleeding and cerebrovascular events were similar to those with alternative access. However, the available data are not adequate to determine how much of the excess mortality seen in patients undergoing alternative access TAVI is caused by the alternative access procedure and how much is caused by excess comorbidity associated with the need for alternative access.

Sex-specific differences — Although females experience more major bleeding and vascular complications, female sex is an independent predictor of lower one-year mortality after TAVI. Thus, sex-specific mortality risk following TAVI is opposite of that following SAVR, for which females have higher mortality risk than males. We suggest including female sex as a factor when weighing the potential risks and benefits of TAVI versus SAVR, since the data suggest that TAVI is superior to SAVR for females with high-risk symptomatic AS.

In the above described four trial meta-analysis with predominantly intermediate- to high-risk patients, there was a borderline significant interaction for sex (p = 0.05), with a significant mortality reduction with TAVI versus SAVR among females (HR 0.68, 95% CI 0.50-0.91) but not among males (HR 0.99, 95% CI 0.77-1.28) [17]. A subgroup analysis of one of the included trials (cohort A of the PARTNER trial in high-risk patients) found that females had lower mortality rates with TAVI compared with SAVR at six months and two-year follow-up, driven by the transfemoral arm [46]. These differences in six-month and two-year mortality rates were not observed in males. (See 'With high surgical risk' above.)

A meta-analysis comparing TAVI outcomes for females and males in 17 studies with a total of 47,188 patients (49.4 percent female) found that females had higher rates of early complications (including bleeding, vascular complications, and stroke/transient ischemic attack), but greater long-term survival compared with males [47]. Survival at 30 days post-TAVI was similar in females and males (93.5 versus 93.8), but was significantly higher in females at one year (84 versus 80.6 percent) and at mean 3.3-year (65.5 versus 61.7 percent) follow-up [46,48,49].

Transcatheter valve type — Data comparing clinical outcomes in patients with AS treated with balloon-expandable versus self-expanding valves are limited. Observational studies of long-term outcomes have not found significant differences related to valve type (eg, United Kingdom registry discussed below). (See 'Data from registries' above.)

The SCOPE trials compared ACURATE neo to SAPIEN and Evolut valves:

●Self-expanding ACURATE neo versus balloon-expandable SAPIEN 3 – The SCOPE 1 trial enrolled 739 patients who were randomly assigned to receive either the ACCURATE neo or SAPIEN 3 valve [50]. Mortality (2 versus 1 percent) and stroke (2 versus 3 percent) rates were similar in the two groups. Moderate or severe prosthetic aortic regurgitation (9 versus 3 percent) and acute kidney injury (11 versus 1 percent) were more common in the ACCURATE neo group.

●Self-expanding ACURATE neo versus self-expanding CoreValve Evolut – The SCOPE 2 trial enrolled 796 patients who were randomly assigned to receive either the ACURATE neo or CoreValve Evolut [51]. Mortality (13 versus 9 percent) and stroke (5 versus 6 percent) rates were similar in the two groups. However, cardiac death at 30 days (2.6 versus 0.8 percent) and 1 year (8.4 versus 3.4 percent) was more frequent in the ACURATE neo group. Moderate or severe aortic regurgitation at 30 days was more also frequent in the ACURATE neo group (10 versus 3 percent). New pacemaker implantations were less frequent with the ACURATE neo device (11 versus 18 percent).

Further study, including an adequately powered study with current valve designs with long-term outcomes is needed to compare balloon- and self-expanding valve types.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cardiac valve disease" and "Society guideline links: Transcatheter aortic valve implantation".)

SUMMARY AND RECOMMENDATIONS

●Surgical aortic valve replacement (SAVR) and transcatheter aortic valve implantation (TAVI) are the mainstays of treatment of severe calcific aortic stenosis (AS), as they improve symptoms and prolong survival. (See 'Symptomatic patients' above.)

●A multidisciplinary heart valve team (including cardiologists, structural valve interventionalists, cardiovascular surgeons, anesthesiologists, and nurses) should collaborate to optimize care for patients with severe AS with indication for valve replacement. The Heart Valve Team should assess an individual patient's life expectancy, frailty, comorbidities, specific anatomy, values, and preferences. (See 'Symptomatic patients' above and 'Individualized risk-benefit assessment' above.)

●If life expectancy with SAVR or TAVI is ≤1 year or the patient's quality of life is unlikely to improve with SAVR or TAVI, palliative therapy with medical management is recommended. (See 'Symptomatic patients' above and "Medical management of symptomatic aortic stenosis".)

●If life expectancy with AVR is >1 year and the patient’s quality of life is likely to improve with SAVR or TAVI, the next step in choosing therapy for symptomatic severe AS is evaluation by the Heart Valve Team of the risk of mortality and morbidity with SAVR (including the Society of Thoracic Surgeons Predicted Risk of Mortality [STS-PROM] and identification of contraindications to SAVR) (algorithm 1). (See 'Symptomatic patients' above.)

•Extreme surgical risk. (See 'TAVI versus medical treatment in inoperable patients' above.)

-For patients with symptomatic severe AS with extreme surgical risk (≥50 percent probability of death or serious irreversible complication) or an absolute contraindication to SAVR and in whom transfemoral TAVI is feasible, we recommend TAVI rather than medical therapy (Grade 1B).

-When transfemoral TAVI is not feasible, the Heart Valve Team should perform an individualized risk-benefit assessment of medical therapy versus alternative access TAVI.

•High surgical risk (ie, STS-PROM >8 with <50 percent probability of death). (See 'With high surgical risk' above.)

-For patients with symptomatic severe AS with high surgical risk in whom transfemoral TAVI is feasible, we recommend transfemoral TAVI (Grade 1B).

-For patients in whom transfemoral TAVI is not feasible, the Heart Valve Team should perform an individualized risk-benefit assessment of SAVR versus alternative access TAVI. (See 'Individualized risk-benefit assessment' above.)

•Intermediate surgical risk (ie, STS-PROM 4 to 8). (See 'In intermediate-risk patients' above.)

-For patients with symptomatic severe AS with intermediate surgical risk in whom transfemoral TAVI is feasible and high risk anatomic features (such as an adverse aortic root, low coronary ostia height, heavily calcified bicuspid aortic valve, and severe left ventricular outflow tract calcification) are absent, we recommend transfemoral TAVI (Grade 1B).

-For patients in whom transfemoral TAVI is not feasible, we recommend SAVR (Grade 1B). In patients with a relative contraindication to SAVR not captured by STS-PROM, an individualized risk-benefit assessment of SAVR versus alternate access TAVI is performed. (See 'Individualized risk-benefit assessment' above.)

•Low surgical risk (ie, STS-PROM <4). (See 'In low-risk symptomatic patients' above.)

-Optimum criteria for choice of intervention in patients with low surgical risk are uncertain. In patients with low surgical risk with all four of the following criteria (age ≥65 years, transfemoral TAVI is feasible, aortic valve is trileaflet, and absence of high risk anatomic features [such as an adverse aortic root, low coronary ostia height, or severe left ventricular outflow tract calcification]), we recommend transfemoral TAVI (Grade 1B). For patients who lack one or more of these four criteria, SAVR is preferred. If a relative contraindication to SAVR is present (unusual in this setting), an individualized risk-benefit assessment of SAVR versus TAVI is performed. (See 'Individualized risk-benefit assessment' above.)

●For patients with asymptomatic severe AS with an indication for valve replacement (eg, presence of left ventricular ejection fraction <50 percent, very severe AS, or fall in systemic blood pressure with exercise), the choice of SAVR versus TAVI is based upon an individualized risk-benefit assessment including the estimated surgical risk (with intermediate or higher estimated surgical risk favoring TAVI and low estimated surgical risk favoring SAVR) and estimated risk of complications with TAVI (with feasibility of transfemoral TAVI and absence of high risk anatomic features favoring TAVI). (See 'Asymptomatic patients' above and 'Individualized risk-benefit assessment' above.)

There is limited evidence comparing SAVR versus TAVI in asymptomatic patients, as these patients were excluded from most of the supporting trials. (See 'Outcomes' above.)

●Individualized risk-benefit assessment of SAVR versus TAVI includes consideration of multiple factors that may occur in combination. Clinical factors influencing the choice between TAVI and SAVR include the patient's values and preferences, age, estimated life expectancy with AVR, as well as presence of concurrent conditions. (See 'Individualized risk-benefit assessment' above.)

•Factors favoring SAVR include presence of another indication for cardiac surgery, lack of femoral access, younger age (longer life expectancy, particularly if patient is a candidate for surgical mechanical valve replacement), and an anatomic feature that would make TAVI high risk (adverse aortic root, low coronary ostia height, heavily calcified bicuspid aortic valve, and severe left ventricular outflow tract calcification).

•Factors favoring TAVI include presence of a relative contraindication for surgery, older age (shorter life expectancy), female sex, and feasibility of transfemoral TAVI.

●Data directly comparing clinical outcomes with balloon-expandable and self-expanding valve are limited; observational studies suggest no significant differences in outcomes among valve types. Further study, including an adequately powered study with long-term outcomes, is needed to compare balloon- and self-expanding valve types. (See 'Transcatheter valve type' above.)

ACKNOWLEDGMENTS

The UpToDate editorial staff acknowledges Catherine M Otto, MD, who contributed as Section Editor to an earlier version of this topic review.

The UpToDate editorial staff also acknowledges William H Gaasch, MD (deceased), who contributed to an earlier version of this topic review.