Management of aortic stenosis: a systematic review of clinical practice guidelines and recommendations

Abstract

Multiple guidelines exist for the management of aortic stenosis (AS). We systematically reviewed current guidelines and recommendations, developed by national or international medical organizations, on management of AS to aid clinical decision-making. Publications in MEDLINE and EMBASE between 1 June 2010 and 15 January 2021 were identified. Additionally, the International Guideline Library, National Guideline Clearinghouse, National Library for Health Guidelines Finder, Canadian Medical Association Clinical Practice Guidelines Infobase, and websites of relevant organizations were searched. Two reviewers independently screened titles and abstracts. Two reviewers assessed rigour of guideline development and extracted the recommendations. Of the seven guidelines and recommendations retrieved, five showed considerable rigour of development. Those rigourously developed, agreed on the definition of severe AS and diverse haemodynamic phenotypes, indications and contraindications for intervention in symptomatic severe AS, surveillance intervals in asymptomatic severe AS, and the importance of multidisciplinary teams (MDTs) and shared decision-making. Discrepancies exist in age and surgical risk cut-offs for recommending surgical aortic valve replacement (SAVR) vs. transcatheter aortic valve implantation (TAVI), the use of biomarkers and complementary multimodality imaging for decision-making in asymptomatic patients and surveillance intervals for non-severe AS. Contemporary guidelines for AS management agree on the importance of MDT involvement and shared decision-making for individualized treatment and unanimously indicate valve replacement in severe, symptomatic AS. Discrepancies exist in thresholds for age and procedural risk used in choosing between SAVR and TAVI, role of biomarkers and complementary imaging modalities to define AS severity and risk of progression in asymptomatic patients.

Introduction

Aortic stenosis (AS) is the most common valvular heart disease in developed countries, and its prevalence is increasing as the population ages,¹ placing an increasing financial burden on healthcare systems.^2,3 Untreated, symptomatic AS is associated with significant morbidity and mortality and has worse survival than many cancers.^4–6 Timely aortic valve intervention returns the mortality curve to that normal for the population at large.⁷

In 2007, the description of a new AS haemodynamic subset, the paradoxical low-flow, low-gradient (LFLG) AS phenotype, eventually led to the recognition of several further haemodynamic subtypes of AS.⁸ Aortic valve area (AVA) and gradients now need to be considered in the context not only of left ventricular ejection fraction (LVEF) but also in that of the flow and flow rate across the stenotic valve.⁹ This adds layers of complexity to the choice and timing of appropriate treatment, in the absence of randomized controlled trials (RCTs) addressing each subset.¹⁰ Therapeutic options for AS have expanded with the advent of transcatheter aortic valve implantation (TAVI), initially reserved for inoperable patients with prohibitive comorbidities, but now shown to be safe and effective even in patients at low surgical risk.¹¹ Long-term transcatheter valve durability remains the main reason for reserved expansion of TAVI in younger patients with AS. Thus, our objective was to perform a systematic review of current guidelines and recommendations from professional societies on surveillance and management of AS in order to inform better clinical decision-making and future research.

Methods

Data sources and searches

We conducted a systematic review of English language guidelines and recommendations for the management of AS in adults. We searched for published guidelines in MEDLINE and EMBASE between 1 June 2010 and 15 January 2021. We also searched the following guideline-specific databases: National Library for Health Guidelines Finder (United Kingdom), Canadian Medical Association Clinical Practice Guidelines Infobase, National Guideline Clearinghouse (United States), and Guidelines International Network International Guideline Library. We also searched websites of organizations relevant to guideline development (Supplementary material online, Appendix Table 1). This systematic review was planned, conducted, and reported in agreement with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations.¹²

Study selection

We included contemporary documents produced by professional organizations that met the Institute of Medicine’s definition of a guideline and referred to adults. If more than one guideline from the same organization existed, we considered the most recent one. Where the most recent guideline was only a focused update, we combined both guidelines.

The MEDLINE search syntax served as a basis for the search strategy (Supplementary material online, Appendix material). The syntax had three elements intersected by the Boolean term ‘AND’: subject headings and free-text terms for the interventions about AS screening, surveillance or management, subject heading and free-text terms for AS, and publication types and title words that cover the clinical practice guidelines.

Data extraction and quality assessment

Titles and abstracts were assessed by two independent reviewers (M.Y.K. and F.R.). Articles were excluded if both reviewers agreed they were ineligible. Discrepancies were resolved by consensus after discussion. Both reviewers performed the final selection for full data extraction.

We used the 23-item Appraisal of Guidelines for Research and Evaluation (AGREE) II instrument to determine the rigour of development for each of the guidelines.^13,14 Two reviewers (V.G. and B.S.) independently rated the items, conforming to the instructions of the AGREE II tool. The average rigour scores were obtained by expressing the sum of the individual scores as a percentage of the maximum possible score. Reproducibility of the two reviewers’ scores was moderate, with an interclass correlation of 0.62 (Supplementary material online, Appendix Table 2). Guidelines were ranked according to their scores. Editorial independence from the funding body, external funding, and disclosure of relationships with the industry by individual guideline group members were also assessed.

Data synthesis and analysis

Two reviewers (M.Y.K. and F.R.) extracted all of the relevant recommendations from the guidelines that had an AGREE II score equal to or greater than 50%. A recommendation matrix was produced.

Role of the funding source

We had no specific funding for this work.

Results

We retrieved 325 titles, of which 29 were potentially eligible. We retained seven guidelines and recommendations on the management of AS after review of the full manuscripts. Five of the seven documents had a rigour score of ≥50% and form the object of this analysis (Figure 1). Table 1 summarizes the selected guidelines along with rigour scores, conflicts of interest.

Figure 1.

Summary of the guideline search and review process. The number of guidelines at each step is indicated. AGREE II, Appraisal of Guidelines for Research and Evaluation II.

Table 1.

Summary of guideline recommendations for aortic stenosis management

Organization—Society	American College of Cardiology/American Heart Association (ACC/AHA)15	European Society of Cardiology (ESC)16	The Society of Thoracic Surgeons (STS)17	Canadian Cardiovascular Society (CCS)18	ACC/AATS/AHA/ASE/EACTS/HVS/SCA/SCAI/SCCT/SCMR/STS19
Document type	Clinical Practice Guidelines	Clinical Practice Guidelines	Clinical Practice Guidelines	Society Position Statement	Appropriate Use Criteria
Country applied	USA	Europe	USA	Canada	USA
Year	2020	2017	2013	2019	2017
AGREE II rigour Score, %	93%	85	77	74	74
Conflict of interest	EI, SCIa,b	EI, SCIa	SCIa	SCIa,b	SCIa
Methods used to evaluate evidence	Systematic review	Systematic review	Systematic review	Systematic review	Selection of relevant guidelines and RCTs
Methods used to formulate recommendations	Formal consensus	Formal consensus	Formal consensus	Agreement of two-thirds of the panel members	Modified Delphi consensus
Consideration of costs	Information from studies on cost considered where available	NR	Information from studies on cost considered where available	NR	When scientific evidence available cost was considered implicitly for AUC determination, but a procedure with moderate efficacy was not scored as more appropriate than a procedure with high clinical efficacy solely due to its lower cost
Definition of severe AS	Vmax ≥4 m/s or mean ΔP ≥40 mmHg AVA ≤1.0 cm2 (or AVAi ≤0.6 cm2/m2) (AVA when transaortic flow rate is abnormal) Very severe AS AVmax ≥5 m/s or mean ΔP ≥60 mmHg Others: LFLG-REF; LFLG-PEF	AVA <1 cm2 and MG >40 mmHg Others: LFLG-REF; LFLG-PEF; NFLG	AVA <1 cm2 (AVA index <0.6 cm2/m2) Vmax >4 m/s, MG >40 mmHg Very severe AS AVA <0.6 cm2, mean gradient >60 mmHg, and jet velocity >5 m/s	NR	Severe HG AS: peak velocity ≥4 m/s or MG ≥40 mmHg, AVA ≤1.0 cm2 (and indexed AVA ≤0.6 cm2/m2) Very severe AS (Vmax ≥5 m/s or MG ≥60mmHg) Low flow: SVi <35 mL/m2 LG: MG <40mmHg (or Vmax <4 m/s) Flow reserve: SVi increase by ≥20% on LDDSE Pseudo-severe AS: AVA with dobutamine >1.0 cm2 and Vmax <4 m/s REF: LVEF <50%
Key diagnostic imaging	Resting TTE (when normotensive) preferred technique for severity assessment	TTE preferred technique for severity assessment	Resting TTE is recommended for the diagnosis and assessment (IB)	Resting TTE preferred technique for severity assessment	Resting TTE preferred technique for severity assessment
Other tests	Low-dose DSE in LFLG-REF (EF <50%) ETT—unmask symptoms, abnormal BP response. Cardiac catheterization for AS severity may be helpful when non-invasive imaging non-diagnostic or discrepancy between clinical and echo evaluation c MSCT—calcium score TAVI planning BNP—Timing of surgery in severe asymptomatic AS (>3 times ULN) CMR in severe LFLG-PEF AS (with optimized blood pressure control)	ETT/ESE—unmask symptoms Low-dose DSE—flow reserve TOE (MV) c MSCT—AV dimensions; calcium score CMR—AV dimensions; fibrosis BNP—timing of surgery in asymptomatic	ETT—unmask symptoms, BP response Low-dose DSE—flow reserve Cardiac catheterization—haemodynamic measurements with infusion of dobutamine for evaluation of patients with low-flow/low-gradient AS and LV dysfunction		ETT—unmask symptoms, abnormal BP response ESE—GM ≥18 mmHg, Low-doseDSE—flow reserve, projected AVA MSCT—Calcium score BNP—timing of surgery in asymptomatic
Surveillance	Every 3–5 years (mild severity Vmax 2.0–2.9 m/s) Every 1–2 years (moderate severity Vmax 3.0–3.9 m/s) Every 6–12 months (Vmax ≥4 m/s)	If asymptomatic severe AS, educate patient about early symptom reporting. Review 6 months with echo, ideally ETT and consider BNP. Yearly—significant Ca2+ and mild and moderate AS 2–3 years—Younger patients with no sig Ca2+ and mild AS	If asymptomatic AS, monitor LVEF, LA >4.5 cm, LVH >180 g/m2, high likelihood of rapid progression (age, calcification, and CAD) (IIbC) Every 6 months for severe AS; 1–2 years for moderate AS; 3–5 years for mild AS (lB)		If asymptomatic severe AS, monitor predictors of symptom onset or of rapid progression (Vmax >0.3 m/s/year, severe valve calcification, BNP, or excessive LV hypertrophy in the absence of hypertension)
Indications for intervention in symptomatic patients	Symptomatic (from history or on ETT) severe AS (1A)d Symptomatic severe LFLG-REF AS (1 B-NR) Severe symptomatic LFLG-PEF AS (1 B-NR)e see section below for asymptomatic patients	Symptomatic severe AS (1B)d Symptomatic severe LFLG-REF AS with contractile reserve (1C)d Severe LFLG-PEF AS (IIaC)e Severe LFLG-REF w/o contractile reserve (IIaC)e	Symptomatic severe AS (IB) Symptomatic or asymptomatic severe AS with LVEF <50% (IC)		Symptomatic truly severe LFLG-REF AS with flow reserve (A) Symptomatic severe LFLG-PEF AS (A) Symptomatic severe LFLG-REF w/o flow reserve (R)
Contraindication for intervention	Existing comorbidities would preclude the expected benefit from correction of AS or where post-intervention expected survival is <12 months, following shared decision-making, including discussion of patient preference and values (1 C-EO)	Severe comorbidities and unlikely to improve QOL or survival	AVR is not useful for the prevention of sudden death in asymptomatic patients with AS who have normal LV systolic function (IIIB) No intervention is recommended in patients with mild or moderate AS unless there are indications for other forms of cardiac surgery Intervention should not be undertaken where comorbidities limit 1-year survival where extreme frailty limits the likelihood of functional recovery	Risk factors for futility if a substantial benefit in terms of QOL or survival is unlikely: Advanced dementia; bedbound, not mobile; cachexia/severe sarcopenia; disability for all/most ADLs; end-stage renal, liver, lung disease, or malignancy	Medical futility Including anticipated life expectancy <1 year, moderate to severe dementia, health status influenced more by comorbidities than by AS
Factors considered for choice of intervention	Optimal individualized treatment decision for TAVI or SAVR based on a number of listed factors or conditions, MDT consensus also based on patient preferences and local expertise and maintenance of competency Choice of proceeding with surgical AVR vs. TAVI based on multiple factors, including the surgical risk, patient frailty, comorbid conditions, and patient preferences and values	Patient cardiac and extracardiac characteristics, risk of surgery judged by Heart Team, scores, feasibility of TAVI and local expertise and outcome data	Preoperative risk of mortality (STS-PROM, EuroSCORE—IIaB) Risk assessment in aged and frail subpopulations judged by qualified Heart Team High risk: >15% risk of death at STS >8 Inoperable: >50% risk of death or irreversible serious postoperative morbidity as assessed by two surgeons	Optimal individualized treatment decision for TAVI or SAVR based on a number of listed factors or conditions, MDT consensus also based on patient preferences and local expertise and maintenance of competency, annual evaluation of quality of TAVI programme	Risk of surgery (STS-PROM), symptoms, uncertainty in distinguishing between severe and pseudo-severe AS, multiple comorbidities, frailty or disability, flow, gradient, and ejection fraction, anatomy, life expectancy, CAD, other valve, ascending aorta pathology, need for major non-cardiac surgery, failing bioprostheses
Surgery	Symptomatic or asymptomatic patients with severe AS and any indication for AVR who are aged <65 or with life expectancy >30 years (1A) Symptomatic AS aged 65–80 years, either surgical or TAVI recommended (as long as no contraindication to transfemoral TAVI) after shared decision-making about balance between expected patient longevity and valve durability (1A) Low risk SAVR defined by STS-PROM <3%, absence of frailty, absence of cardiac or other major organs compromise and absence of procedure-specific impediments [tracheostomy, heavily calcified (porcelain) ascending aorta, chest malformation, arterial coronary graft adherent to posterior chest wall, and radiation damage] Futility: STS >15%, life expectancy <1 year, poor candidate for rehabilitation	Low surgical risk (STS or EuroSCORE II <4% or logistic EuroSCORE I <10%). No frailty, porcelain aorta, sequelae of chest radiation (1B) Other patient specific factors (age <75 years, unfavourable femoral access, unsuitable annulus size for TAVI), bicuspid aortic valve. Other cardiac conditions requiring intervention (endocarditis, need for CABG, significant other valve disease, aortic aneurysm)	Severe AS at the onset of symptoms (IB) Regardless of symptoms, with the identification of LV systolic dysfunction (LVEF<50%) (IC) AVR is recommended in patients with severe AS who are scheduled to undergo CABG, surgery on other cardiac valves, or surgery on the aortic root or ascending aorta (IC)	Conditions that should be considered for recommending SAVR: Aortic root anatomy unfavourable for TAVI (excessive calcification, annulus size out of range) Advanced AVB, especially RBBB Non-femoral access required Congenital bicuspid valve Risk of coronary obstruction or coronary access concerns Pure AR Multivalvular disease Aortic aneurysm or dissection Endocarditis	Severe symptomatic AS and other valve or ascending aortic pathology or myectomy STS-PROM <15% (extreme high risk) Prominent basal septal hypertrophy with flow acceleration and narrowing in the LVOT Small annular size Small failing bioprostheses (≤19 mm)
TAVI	Symptomatic severe AS patients aged >80 years or if life expectancy <10 years for younger patients and no anatomical contraindication for transfemoral TAVI—TAVI preferred over surgical AVR (IA) Symptomatic AS aged 65–80 years, either surgical or TAVI recommended (as long as no contraindication to transfemoral TAVI) after shared decision-making about balance between expected patient longevity and valve durability (IA) Futility: STS >15%, life expectancy <1 year, poor candidate for rehabilitation	Surgical risk not low and Heart Team deems patients not suitable for SAVR (1B) Favoured in elderly (≥75) with TF access	Transfemoral, transaortic, trans-axillary, or transapical AVR with balloon expandable valve can be considered in patients who are operative candidates and have a predicted surgical mortality >15% and an STS score >10% by two independent surgical assessments TAVI should be performed by a multidisciplinary cardiovascular and cardiac surgery team	Conditions that should be considered for recommending transfemoral TAVI: Intermediate or high risk of surgical morbidity and mortality (STS score ≥3) Elderly (≥75 years), frailty, limited mobility Small annulus (prosthesis size ≤21 mm) Mediastinal anatomy unfavourable for surgery (porcelain aorta, previous thoracotomy, patent grafts, hostile root) ViV-TAVI as an alternative to reoperation in failed bioprosthetic valve	High surgical risk (STS-PROM ≥8%) High- or extreme-risk patients due to: Multiple comorbidities, frailty or disability, and anticipated life expectancy >1 year (A) Anatomy: porcelain aorta or hostile chest (A) specific comorbidities: oxygen-dependent lung disease (A), dialysis (A, M), cirrhosis with MELD >14 (A)
BAV	Considered as bridge to AVR or TAVI in critically ill patients with severe AS (IIb C-EO)	Bridge to TAVI or SAVR haemodynamically unstable Severe AS (IIb C)e Diagnostic Tx severe AS but SOB may be due to lung, reversible myocardial or organ dysfunction may improve with BAV (IIb C)e	Bridge to AVR in haemodynamically Unstable severe AS Where immediate AVR is not feasible (IIaC) BTD if contraindication to AVR (IIbC) but potential bridge to TAVI or SAVR in the future (IIaC) BTD when contribution of AS to symptoms such as COPD or poor LV function, remains unclear (IIaC) Urgent major non-cardiac surgery (IIbC)	NR	BDT in truly severe LFLG-REF AS (M) and NFLG-PEF AS (M) Failing bioprostheses (R) Life expectancy <1 year or moderate to severe dementia (M)
SAVR in asymptomatic severe AS	LVEF <50% due to severe AS (1 B-NR)d, abnormal ETT (IIa B-NR)e Undergoing other cardiac surgery (1 B-NR) or Low surgical risk and either Vmax >5m/s, (IIa B-NR) Rate of Vmax progression ≥0.3 m/s/year (IIa B-NR)e or decreased exercise tolerance BP drop on ETT (IIa B-NR) If severe AS (1 B-NR)d, or moderate AS (IIb C-EO) at time of cardiac surgery for other indication	LVEF <50% due to severe AS (1C)d, symptomatic on ETT (1C)d, Drop in BP with ETT (IIaC)e OR low surgical risk and either Vmax >5.5m/s, severe valve Ca2+ and rate of Vmax progression ≥0.3 m/s/year, >3× predicted BNP, invasive PASP >60 mmHg due to AS (IIaC)e If severe AS (1C)d, moderate AS at time of CABG or aorta or other valve surgery if Heart Team agrees (IIaC)e	Asymptomatic patients with extremely severe AS (AVA <0.6 cm2, MG >60 mmHg, and Vmax >5.0 m/s) when expected operative mortality is <1%. AVR may be considered for asymptomatic patients with severe AS and abnormal response to exercise (e.g. asymptomatic hypotension) (IIbC) AVR may be considered for adults with severe asymptomatic AS if there is a high likelihood of rapid progression (age, calcification, and CAD) or if surgery might be delayed at the time of symptom onset (IIbC) AVR may be considered for asymptomatic patients with extremely severe AS when the patient’s expected operative mortality is less than 1% (IIbC)	NR	Asymptomatic severe AS (M) Asymptomatic severe AS and LVEF<50% (A) High-risk profession (e.g. airline pilot) or lifestyle (e.g. competitive athlete) (A) Anticipated prolonged time away from close medical supervision (A) Undergoing another cardiac surgery or ascending aortic surgery (A) Very severe AS (Vmax ≥5 m/s or mean gradient ≥60 mmHg) (A) ≥1 predictor(s) of symptom onset or of rapid progression (Vmax >0.3 m/s/year, severe valve calcification, elevated BNP, or excessive LVH in the absence of hypertension) (A) Symptomatic severe HG AS With associated CAD (A)
Medical therapy	No evidence of specific medical treatment for AS	No evidence of specific medical treatment for AS	No evidence of specific medical treatment for AS	No evidence of specific medical treatment for AS	No evidence of specific medical treatment for AS

ACC/AATS/AHA/ASE/EACTS/HVS/SCA/SCAI/SCCT/SCMR/STS, American College of Cardiology, American Association for Thoracic Surgery, American Heart Association, American Society of Echocardiography, European Association for Cardio-Thoracic Surgery, Heart Valve Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Computed Tomography, Society for Cardiovascular Magnetic Resonance, and Society of Thoracic Surgeons; AGREE II, Appraisal of Guidelines Research and Evaluation II; AR, aortic regurgitation; AVA, aortic valve area; BAV, balloon aortic valvuloplasty; BDT, bridge to decision; CAD, coronary artery disease; CEA, cost-effectiveness analysis; CKD, chronic kidney disease; CVD, cardiovascular disease; DM, diabetes mellitus; DSE, dobutamine stress echocardiography; EI, editorial independence declared; FIP, funding by industrial partner reported; LFLG, low-flow low-gradient; MG, mean gradient; NFLG, normal-flow low-gradient (valve area <1 cm², mean gradient <40 mmHg, ejection fraction ≥50%, SVi >35 mL/m²); NR = not reported; PEF, preserved ejection fraction (valve area <1 cm², mean gradient <40 mmHg, ejection fraction ≥50%, SVi ≤35 mL/m²); REF, reduced ejection fraction [valve area <1 cm², mean gradient <40 mmHg, ejection fraction <50%, stroke volume index (SVi) ≤35 mL/m²]; SCI, statement about conflicts of interest of group members present; TAVI, transcatheter aortic valve implantation; TF, transfemoral; TTE, transthoracic echocardiography; ULN, upper limit of normal; ViV, valve in valve.

^aRelationship with industry is reported by any group member.

^bA group member is reported recused when a relevant area is under discussion.

^cOther key test before TAVI or when severity unclear.

^dIndicated.

^eConsidered.

Level of evidence: A = Data derived from multiple randomized clinical trials or meta-analysis. B = Data derived from a single randomized trial or nonrandomized studies. C = Only consensus opinion of experts, case studies or standard of care.

Recommendation class: Class I = Benefit ⋙ risk. Class IIa = Benefit ≫ risk. IIb benefit ≥ risk.

ACC/AATS/AHA/ASE/EACTS/HVS/SCA/SCAI/SCCT/SCMR/STS 2017 appropriate use criteria.

A, appropriate care; M, may be appropriate care; R, rarely appropriate care.

The application of class of recommendation and level of evidence characterization of STS clinical practice guidelines is according to those recommended by ACCF/AHA: Level of Evidence (LoE) A, B (B-R = randomized study, B-NR = non-randomized), C (C-LD = limited data, C-EO = consensus of expert opinion).

Class of recommendations: I, conditions for which there is evidence for and/or general agreement that the procedure or treatment is beneficial, useful, and effective; Class II: conditions for which there is conflicting evidence and/or a divergence of opinion about the usefulness/efficacy of a procedure or treatment; IIa, weight of evidence/opinion is in favour of usefulness/efficacy; IIb, usefulness/efficacy is less well established by evidence/opinion; III, conditions for which there is evidence and/or general agreement that the procedure/treatment is not useful/effective and in some cases may be harmful.

Areas of agreement

Definition of severe AS.

Four of the five documents provided a definition of severe AS based on a high gradient. Three of them discussed other types of severe AS: LFLG with reduced ejection fraction (rEF) and LFLG with preserved ejection fraction (PEF). Three of the recommendations also defined a ‘very severe AS’ class [aortic valve (AV) max ≥5 m/s or mean gradient ≥60 mmHg].

Diagnostic tests

There was consensus that transthoracic echocardiography (TTE) with Doppler is the key diagnostic test, with the use of AVA when transaortic flow rate is not normal. Four of the guidelines also mentioned the use of other tests for assessment of severe AS and recommended exercise testing (ETT or exercise stress echo) to unmask symptoms in those who may appear to be asymptomatic. Low-dose dobutamine stress echo was recommended in three recent guidelines in the presence of rEF and to assess flow reserve. The multi-societal appropriateness of use criteria (AUC) document was the sole report recommending calculation of projected valve area during dobutamine stress echocardiography (DSE) to guide valve replacement in patients with severe left ventricular (LV) dysfunction and low stroke volume index.

Indications and contraindications for intervention

Four guidelines identified symptomatic severe AS as a clear indication for intervention. There was also agreement that patients with a reduced LVEF and severe AS should undergo intervention. Three of the guidelines elaborated further and included LFLG with rEF while two mentioned LFLG-PEF. For severe asymptomatic AS, three of the guidelines recommend surgery when LVEF is ≤50% when cardiac surgery is indicated for another reason, or with low surgical risk (<1%) and high likelihood of progression such as suggested by an abnormal blood pressure response on ETT or in the context of very severe AS (Vmax ≥5/≥5.5).^{16,17,19–21}

There was agreement that intervention is contraindicated in patients with severe comorbidities and when intervention is unlikely to improve quality of life or when life expectancy is <1 year. The importance of the multidisciplinary team (MDT) in this process was strongly highlighted.

Choice of intervention

Where the surgical risk is low or intermediate or in the presence of other cardiac conditions that require intervention (multivalvular disease, multi-vessel coronary artery disease (CAD) requiring coronary artery bypass graft (CABG), endocarditis or thoracic aortic aneurysm), surgical aortic valve replacement (SAVR) is preferred. Younger age and presence of bicuspid AV also favoured SAVR.

Balloon aortic valvuloplasty was mentioned in four of the five guidelines as a bridge to TAVI or SAVR in patients who are haemodynamically unstable or where it is not clear if dyspnoea is due to a respiratory cause or to the AS [European Society of Cardiology (ESC) and Society of Thoracic Surgeons (STS)]. There is general consensus that choice between TAVI and SAVR should be based on multiple factors, including surgical risk, patient frailty, comorbid conditions, and patient preferences and values, in order to make an optimal, individualized decision on treatment. There is also specific mention of the central role of the MDT or heart team involvement in four of the guidelines.

Guidelines agree that there is no specific medical therapy to halt or slow the progression of AS. None of the recommendations currently mention population-based screening.

Surveillance frequency

Three of the five guidelines agreed that surveillance should take place 6-monthly in asymptomatic severe AS, while American College of Cardiology/American Heart Association (ACC/AHA) recommends a range of 6–12 months.

Areas of disagreement

Choice of TAVI over SAVR

Although there is general agreement that the decision for TAVI should be undertaken by an MDT or heart team, the criteria used to make the decision are inconsistent. Both ESC and Canadian Cardiovascular Society (CCS) favour TAVI in those ≥75 years or older. The 2020 ACC/AHA favour TAVI in those over 80 years of age or life expectancy of <10 years. They recommend that either SAVR or TAVI can be considered in those aged between 65 and 80 years (as long as there are no contraindications to transfemoral TAVI) after shared decision-making about the balance between expected patient longevity and valve durability. Although guidelines recommend quantifying the risk of SAVR and where this is intermediate or high then TAVI is preferred, there is no agreement on what constitutes intermediate or high risk. Most of the guidelines recommend using the STS-PROM score but the cut-offs are variable (3–15% depending on the source). Only some of the guidelines identify specific anatomical or clinical features unfavourable for surgery (e.g. porcelain or ‘hostile’ aorta, previous thoracotomy),^16,19,20 and specifically following coronary artery bypass surgery with patent grafts.

The STS guideline is the only one that specifically mentions that aortic valve replacement (AVR) is not indicated for the prevention of sudden death in asymptomatic severe AS patients who have normal LVEF, although the oldest document included in this review.

Imaging markers and biomarkers

There is no agreement on biomarkers or on other imaging beyond TTE for further assessing the severity of AS. Three of the guidelines mention brain natriuretic peptide (BNP) levels to guide timing of surgery in asymptomatic patients, with the ACC/AHA and the ESC guideline gives a quantitative cut-off (serum level > times normal).

Multi-slice computed tomography is recommended in three of the guidelines for calcium scoring in three of the guidelines and for AV dimensions; two of the guideline mentioned cardiac magnetic resonance (CMR) for the assessment of AV dimensions and one for the assessment of myocardial fibrosis.

Surveillance frequency

In mild AS ACC/AHA recommends 3–5 years of follow-up whereas ESC recommends yearly surveillance for those with ‘significant calcification’ and 2–3 yearly for younger patients with no significant calcification.

For moderate AS, two of the guidelines recommend 1–2 yearly follow-up. The ESC guideline stipulates yearly follow-up in the presence of significant calcification, although it does not specify the interval in the absence of calcification.

Discussion

We identified seven clinical practice guidelines and recommendations, of which five were rigourously developed, on the management of patients with AS. There is consensus on the definition of severe AS and low-flow/gradient types, on indications and contraindications for intervention in symptomatic severe AS, on the importance of MDT for individualized management, and on surveillance intervals in asymptomatic severe AS. Discrepancies exist in age and surgical risk cut-offs for deciding on a specific type of intervention (SAVR vs. TAVI), the use of biomarkers (e.g. BNP) and complementary multimodality imaging for decision-making in asymptomatic patients and in surveillance intervals for non-severe AS. Recommendations on the role of minimally invasive surgery and non-transfemoral TAVI approaches are limited. None of the recommendations mention screening for AS in the general population, or optimal transfusion thresholds for red blood cell transfusion after TAVI.²²Figure 2 provides a summary of the areas of agreement, disagreement and potential gaps in evidence requiring future clarification.

Figure 2.

Summary of clinical practice guidelines and recommendations on aortic stenosis management. AS, aortic stenosis; HFrEF, heart failure with reduced ejection fraction; LV, left ventricle; MDT, multidisciplinary team; SAVR, surgical aortic valve replacement; TAVI, transcatheter aortic valve replacement.

Impaired LVEF

Guidelines recommend valve replacement for symptoms and/or impaired LV systolic function with evidence of flow reserve during low-dose DSE and exclude the pseudo-severe phenotype for which the role of AVR is not clear. Projected valve area at normal flow rate provides additional diagnostic and prognostic information in patients with low-flow low-gradient AS, on a par with resting transvalvular flow rate.^23–26 More evidence is needed to assess whether resting flow rate and projected valve area should be incorporated into clinical algorithms.

Moderate AS

Currently, guidelines do not recommend AVR in patients with LV systolic dysfunction and moderate AS at rest and/or on low-dose dobutamine stress echocardiogram. Nevertheless, there is excess mortality associated with both moderate and severe AS during long-term follow-up, with a threshold of increased risk of long-term all-cause and cardiovascular death for a mean AV gradient of 20 mmHg and peak AV velocity of 3 m/s.²⁷ Hence, it has been hypothesized that relief of AS earlier in the course of the disease might be favourable, especially in the setting of heart failure (HF) with reduced ejection fraction (HFrEF). The TAVI UNLOAD trial (NCT02661451) will randomize 600 patients with moderate AS and HFrEF to guideline-directed HF therapy alone vs. guideline-directed HF therapy plus TAVI in order to evaluate whether early TAVI reduces readmissions and improves LV systolic function, functional status, exercise capacity and survival.^28,29

Asymptomatic patients

While in symptomatic patients AVR relieves symptoms and prolongs life, decisions are difficult in asymptomatic patients: one needs to balance the risks of intervention and of an artificial valve against those of continued watchful waiting, including that of sudden cardiac death. The progression to symptoms is inevitable, hence the more pertinent question is not ‘if’ the valve should be replaced, but ‘when’.³⁰ If symptom status is unclear, including in the elderly patients where symptoms may be difficult to ascertain due to comorbidities or mobility impairment, exercise testing/exercise stress echo, serum BNP and predictors of rapid progression (rapid change in peak jet velocity, disproportionate LV hypertrophy (LVH) and severe valve calcification) become important elements in decision-making.^31,32 Interestingly, the American multi-societal AUC documents introduced high-risk profession (e.g. airline pilot), lifestyle (e.g. competitive athlete) or anticipated inaccessibility of close medical supervision as clinical circumstances where referral to AVR would be appropriate.

Recently, the RECOVERY trial demonstrated a lower incidence of operative mortality, cardiovascular or all-causes death at 6 years in patients with very severe AS who had early AVR compared to those who received conservative care.³³ The RECOVERY trial heralds a potential future indication for early surgery in patients with very severe AS, despite differences in the definition of very severe AS (AVA ≤ 0.75 cm² with either peak aortic jet velocity of ≥4.5 m/s or mean transaortic gradient of ≥50 mmHg), compared with European and American guidelines.

Imaging markers and biomarkers

Ongoing randomized trials of early TAVI in patients with asymptomatic severe AS (AVATAR, NCT02436655; EVOLVED, NCT03094143; ESTIMATE, NCT02627391; and EARLY TAVI, NCT03042104) will direct future guidelines and provide new evidence on the role of multi-modality imaging biomarkers of subclinical LV decompensation and fibrosis, such global longitudinal strain, T1 mapping indices and late gadolinium enhancement.^34–36 Notably, multimodality non-invasive cardiac imaging provides the capability to image directly both the aortic valve and the myocardium, to monitor disease activity and progression, and to provide efficacy endpoints in ongoing trials of novel therapies such as alternative lipid-lowering approaches (EAVall—Early Aortic Valve Lipoprotein(a) Lowering Trial—NCT02109614), anti-calcification therapies (SALTIRE II—Study Investigating the Effects of Drugs Used to Treat Osteoporosis on the Progression of Calcific Aortic Stenosis—NCT02132026) and novel antifibrotic strategies.³⁷

Alternatives to transfemoral TAVI

Beyond guidance on the choice between SAVR and TAVI, no clear recommendations are generally available for minimally invasive SAVR,³⁸ or for specific vascular approaches supplanting non-transfemoral TAVI. The STS guideline includes a detailed section on the approaches and incisions for performing aortic valve interventions. The transfemoral route is generally the preferred approach, with alternative access options including transaortic, trans-subclavian/trans-axillary and transapical access, but also trans-carotid, trans-caval, and antegrade aortic, mostly restricted to experienced operators.^39,40

Cardiac amyloidosis

With advances in cardiac imaging the reported prevalence of concomitant AS and transthyretin cardiac amyloidosis—ATTR-CA—ranges between 6% and 25% in different age cohorts. ATTR-CA signifies higher all-cause mortality in elderly patients with AS and the degree of maximum LV wall thickness (LVWT) appears to be a major prognostic determinant in patients with dual pathology, regardless of age, LVEF, and type of aortic valve intevention.⁴¹ So far there are no recommendations on whether patients with AS should be systematically screened for CA.⁴²

Risk scores

All documents included in this systematic review endorse the use of risk-assessment models. EuroSCORE, EuroSCORE II, and STS-PROM are recommended to predict surgical outcomes, aid in treatment selection, patient counselling, comparison of postoperative results, and quality improvement.³⁶ Beyond well-known limitations of both EuroSCORE and STS-PROM risk scores, including need for periodic recalibrations and lower accuracy in TAVI procedures, we found inconsistencies across STS recommendations for definition of intermediate, high, or extremely high-risk thresholds that underpin treatment selection, which has possible implications on patient outcomes, depending on which set or recommendations directs clinical practice in a specific geographical area.

Most recent 2020 ACC/AHA guidelines recommend that the decision to intervene, as well as the type of procedure recommended, should be based on an assessment of patient-, procedure-, and institution- or operator-specific short-term risks and long-term benefits, but should also rely on several prognostic factors not captured in currently available risk scores, such as aortic arch atherosclerosis with protuberant lesions, severe mitral regurgitation (MR) or tricuspid regurgitation, low-lying coronary arteries, basal septal hypertrophy, valve morphology (e.g. bicuspid or unicuspid valve) and extensive LV outflow tract calcification for TAVI, or porcelain aorta for SAVR.

With regards to very high-risk setting, both SAVR and TAVI should be considered as futile in individuals presenting with STS >15%, life expectancy <1 year and/or deemed as poor candidate for rehabilitation.

Novel risk assessment models such as the Edwards score, based on the STS/ACC TVT registry, or the German Aortic Valve Score, based on German aortic valve registry (GARY) may improve selection of TAVI patients and predict procedural risk, although long-term data, refinement and comparative effectiveness analyses are required.^43–45 The 2020 ACC/AHA guidelines endorsed the STS/ACC TAVR in-hospital mortality risk calculator, developed from the abovementioned STS/ACC TVT registry, and intended to be used by physicians at the point-of-care to evaluate a patient’s potential risk of death while in the acute care setting from the TAVI procedure. This risk-adjusted mortality estimate is recommended to be used for guidance in the overall conversation about the TAVI procedure, and not as a recommendation for or against any medical procedure.

Surveillance

In patients with mild AS, yearly history and physical examination and TTE every 3–5 years is appropriate according to ACC/AHA recommendations, while the ESC recommend annual assessment of patients with mild or moderate AS and significant calcium burden, or 2–3 yearly in younger patients with mild AS and no calcification with no specific mention for those with bicuspid aortic valves.^16,20 Variations between different recommendations may lead to inefficient resource utilization or, conversely, to inadequate monitoring.⁴⁶ Cost-effectiveness and outcome studies are needed to define the optimum frequency of monitoring.⁴⁷ Patients with severe AS should be examined twice yearly or sooner if they develop symptoms that are potentially attributable to AS.^16,17,20

Screening

Recommendations for or against population screening for AS or family screening for those with bicuspid aortic valves is not mentioned in any of the guidelines—likely due to limited evidence on cost-effectiveness and feasibility—but may justify further research given the aging population and poor outcome related to late valve intervention.²⁷ In order to mitigate the impact of AS on society and to curb healthcare expenditure the first challenge is to improve the timely diagnosis of AS in a cost-effective manner, for instance through systematic targeted screening in primary care, in residential care facilities and/or during vaccinations in the elderly, perhaps using point-of-care hand-held ultrasound scaners.^48,49

Cardiac catheterization

Cardiac catheterization for haemodynamic assessment of AS is seldom performed today, when most uncertainties about the severity of the valve lesion can be resolved with judicious use of non-invasive imaging.²⁸ Both the 2017 AHA/ACC and the ESC guidelines recommend right heart catheterization if there is a suspicion of pulmonary arterial hypertension.^16,20 The 2020 AHA/ACC guidelines recommend invasive haemodynamic assessment—to be performed with meticulous attention to detail by skilled operators with knowledge and expertise in valvular heart disease assessment—in cases of inconclusive non-invasive measurements or discordance between non-invasive tests and clinical findings,⁵⁰ and highlight the usefulness of invasive aortography for the quantification of aortic regurgitation.¹⁵

Patients with AS often have concomitant CAD that may require myocardial revascularization, and coronary angiography before valve intervention is indicated by all guidelines. There is agreement in recommending invasive coronary angiography as the preferred option, but computed tomography coronary angiography is an alternative in patients with low pre-test probability of CAD,¹⁵ as it has excellent negative predictive value, or in patients in whom invasive angiography is technically difficult or not feasible.¹⁶

Coronary revascularization

If SAVR is recommended, surgical experience indicates improved outcomes with concomitant CABG for patients who have significant CAD, defined as >50% stenosis of an epicardial coronary artery (ACC/AHA 2020). With TAVI, there is consensus that revascularisation of significant coronary artery lesions by percutaneous coronary intervention (PCI) should be performed, but no RCT-level data regarding its timing relative to the valve procedure. ACC/AHA guidelines state that usual practice is to perform PCI before TAVI in patients with significant CAD, but also recommend that CABG (in conjunction with SAVR) should be preferred over PCI for patients with anatomically complex left main stem disease.¹⁵

TAVI in bicuspid AS

The pivotal RCTs that established TAVI as an effective and safe treatment did not include patients with bicuspid AS. Moreover, such patients tend to be younger than the typical TAVI patient population, meaning that TAVI valve durability—still not fully proven—becomes even more important. In the absence of RCTs we are left with registry data, which—so far—suggest an acceptable safety profile and results of TAVI in bicuspid AS,⁵¹ even in the low surgical risk setting.⁵²

Cerebral protection during TAVI

Stroke and transient ischaemic attack (TIA), as well as asymptomatic cerebral embolism, are potential complications of TAVI. In PARTNER 1B (TAVI vs. medical treatment in high surgical risk AS patients) medically treated patients had a 30-day incidence of stroke/TIA of 1.7%, and a 1-year incidence of 4.5%, compared with 6.7% and 10.6%, respectively.⁵³ In PARTNER 3 (low surgical risk) stroke rates were identical between TAVI and SAVR at 30 days (3.4% in both arms), and very similar at one year (4.1% vs. 4.3%).⁵⁴

Various cerebral protection devices have been developed with the aim of preventing embolization of atheroma or calcific debris to the brain, either by retaining the debris (filters), by ‘re-routing’ the embolized material away from the cerebral circulation (deflectors) or by combining the two in a ‘total protection device’.^55,56 The evidence so far—in the absence of large RCTs—does not support the routine use of these devices during TAVI. The PROTECTED TAVI (NCT04149535) RCT may clarify this area.

Study limitations

There are some limitations that could bias our findings and require attention. Firstly, only guidelines developed by Western national or international medical organizations were reviewed, but we controlled for selections bias with a comprehensive search strategy and rigourous selection by two independent researchers. Secondly, detailed recommendations on multi-modality imaging for the diagnosis of valvular heart disease, endocarditis prophylaxis, long-term follow-up, and antithrombotic management were not covered in this review as dedicated analyses are available in specific clinical practice guidelines. Thirdly, although we considered the guideline development process, wet did not assess the clinical validity of the individual recommendations as it was beyond the scope of this review.

Conclusions

Contemporary guidelines for the diagnosis and management of AS achieve consensus on the importance of MDT involvement for individualized treatment decisions and unanimously indicate valve replacement in severe, symptomatic AS. There are significant discrepancies in thresholds for age and procedural risk used in choosing between SAVR and TAVI, with variation in the use biomarkers and complementary imaging modalities utilized to define AS severity. Further research in these areas as well as on the role of population screening appears justified.

Supplementary material

Supplementary material is available at European Heart Journal – Quality of Care and Clinical Outcomes online.

Acknowledgements

We would like to thank Dr Francesca Pugliese for her constructive feedback that helped improve the manuscript.

Funding

C.A.A.C. is supported by the National Institutes of Health [NIH (HL134885)], the Mayo Clinic Foundation for Medical Education and Research, and the Paul and Ruby Tsai Foundation. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official view of the NIH.

Conflict of interest: none declared.