Article Text

Early surgery versus watchful waiting for asymptomatic severe aortic valve stenosis: a hot topic for the past 20 years
  1. Sylvestre Marechaux1,2,
  2. Christophe Tribouilloy2,3
  1. 1Cardiology Department, GCS-Groupement des hôpitaux de l'institut Catholique de Lille, Université Catholique de Lille, Lille, France
  2. 2INSERM U 1088, Université de Picardie, Amiens, France
  3. 3Cardiology Department, Centre Hospitalier Universitaire d'Amiens, Amiens, France
  1. Correspondence to Professor Christophe Tribouilloy, Department of Cardiology, Avenue René Laënnec, Amiens 80054, Cedex 1, France; tribouilloy.christophe{at}

Statistics from

Severe aortic stenosis (AS) is currently defined as an aortic valve area (AVA) <1.0 cm2 and/or mean trans-aortic pressure gradient >40 mm Hg and/or peak aortic jet velocity (Vmax) >4 m/s. Only patients with severe AS associated with symptoms or left ventricular ejection fraction (LVEF) <50% present an European Society of Cardiology class I indication for aortic valve replacement (AVR),1 based on the findings of Ross's and Braunwald's landmark report showing a dramatic increase in mortality after symptom onset in patients with AS while ‘operative treatment was deemed to be the most common cause of sudden death in asymptomatic AS patients’. The annualised rate of sudden death is estimated to be around 1% per year in asymptomatic patients, which must be weighed up against the operative mortality of AVR (1%–3% in patients aged <70 years and 3%–8% in older patients). In addition, the native valve is usually replaced by a prosthetic valve, which is associated with specific life-threatening complications (thrombosis, endocarditis, need for reoperation). In contrast, it has been suggested that some patients with severe asymptomatic AS may be operated at an excessively advanced stage of the disease, at which myocardial impairment is at least partially irreversible, consequently resulting in a higher risk of mortality and heart failure (HF). This debate remains unresolved, as some experts suggest early surgery to spare the left ventricle (LV) and improve long-term survival, while others advocate AVR only after onset of symptoms.

The slowly progressive nature of AS combined with the relatively advanced age of the population affected by this disease predispose to under-reporting and/or underestimation of symptoms. Thus, a recent series based on cardiopulmonary exercise testing reported a 28% rate of ‘false asymptomatic AS patients’. Exercise testing (ie, exercise testing, exercise echocardiography, cardiopulmonary exercise testing) should be performed whenever possible to detect patients who are not ‘truly’ asymptomatic and who should undergo AVR in the presence of a reasonable operative risk.1

In addition to the class I indication for AVR (symptoms or LVEF<50%), AVR should be considered in asymptomatic patients with a fall in blood pressure during exercise, with very severe AS defined by Vmax>5.5 m/s or severe valve calcification and a rate of Vmax progression of ≥0.3 m/s per year (class IIa recommendations).1 Patients without these predictive factors should be managed conservatively with follow-up every 6 months to detect changes in symptoms, an acceptable frequency compared with the estimated 1% per year risk of sudden death. However, early surgery is not formally contraindicated by current guidelines, as indicated by the absence of a class III recommendation for early surgery in asymptomatic patients with severe AS.1 In this issue, Wei et al2 present a meta-analysis of four retrospective observational studies enrolling asymptomatic patients with asymptomatic severe AS. This meta-analysis compared overall mortality between patients who underwent AVR after symptom onset with those who underwent AVR before symptom onset. Patients who developed symptoms but who did not undergo AVR were excluded from the analysis. No significant difference (p=0.1) in terms of overall mortality was observed between the two approaches. Similarly, no significant difference in terms of cardiac mortality and sudden cardiac death was observed between the two approaches in the three studies reporting these end points. This meta-analysis does not support a strategy of AVR before onset of symptoms in asymptomatic patients with severe AS. Interestingly, another meta-analysis on exactly the same topic was published several months ago.3 On the basis of four retrospective studies (two of which were included in Wei et al's meta-analysis), the authors concluded that asymptomatic patients with severe AS have a 3.5-fold higher rate of all-cause death with a watchful-waiting strategy compared with an early AVR strategy. The discrepancy between the results of these two meta-analyses can be explained by differences in methods and the studies selected, which were all subject to possible selection bias and were predominantly retrospective. It is noteworthy that in the conservative group of the largest report included in both meta-analyses, more than 40% of patients had not undergone AVR 2 years after inclusion despite a class I indication and almost 50% of patients who developed symptoms did not undergo AVR.4 Two (weight: 56.1%) of the four studies from Wei et al's meta-analysis compared an initial AVR approach at the time of diagnosis with a conventional approach; the other half of the studies included asymptomatic patients with severe AS, some of whom underwent AVR during follow-up despite remaining asymptomatic, thereby constituting the comparator group. In Généreux et al's meta-analysis, the weight of the studies comprising a group of patients undergoing initial AVR was higher (68.6%) than in Wei et al's meta-analysis (56.1%). The discordant conclusions drawn by these two meta-analyses clearly illustrate the limitations of such meta-analyses and their hypothesis-generating nature. Consequently, no recommendations for the management of asymptomatic patients with severe AS in routine practice can be derived from these meta-analyses.

Prospective randomised control trials comparing early surgery with a symptom-driven approach are therefore required to resolve this issue. The AVATAR (Aortic Valve replAcemenT versus conservative treatment in Asymptomatic seveRe aortic stenosis) trial is designed to assess outcomes (global composite outcome comprising: all-cause death, acute myocardial infarction, cerebrovascular insult and unplanned hospitalisation for HF requiring intravenous treatment) in 312 asymptomatic patients with severe AS having normal exercise testing randomised to either elective early AVR or watchful waiting.5 The hypothesis is that elective AVR will reduce the primary outcome compared with medical management and AVR after symptom onset. Importantly, a large set of echocardiographic, laboratory, cardiopulmonary exercise testing and cardiac MRI (cMRI) data will be collected, allowing the identification of predictors of adverse events in the ‘watchful waiting’ arm. Another randomised prospective study on the same topic will be initiated in the near future in France (ESTIMATE-Early Surgery for paTIents with asympoMAtic aortic sTEnosis- study).

However, the prognosis on conservative management as well as the operative risk of asymptomatic patients with severe AS is not uniform. In view of the surgical risk of AVR and the complications of the prosthesis that may affect the patient's outcome, it seems reasonable to hypothesise that only a subset of ‘high-risk patients’ would benefit from early surgery. In addition, transcatheter AVR (TAVR) in asymptomatic severe AS is clearly not currently recommended in low- surgical-risk patients. Ongoing trials should elucidate the benefit of TAVR in low-surgical-risk patients in the near future, as the procedural risks of TAVR have been dramatically reduced over the past 5 years. In Kang et al's retrospective report, comparing the outcome of an early surgery approach with a conventional approach,6 Vmax >5 m/s was a predictor of cardiac mortality in the conventional treatment arm. Similarly, patients with an AVA of ≤0.6 cm2 had an increased risk of all-cause mortality compared with patients with severe AS but an AVA of 0.6–1 cm2.7 In addition, the rate of haemodynamic progression of AS, which has been related to the onset of new symptoms and death within 2 years, varies widely from one patient to another. Although older age, coronary artery disease, severe stenosis at baseline and degree of calcifications assessed by echocardiography or preferably by cardiac CT have been identified as predictors of rapid progression of AS, this prediction remains difficult in individual patients. Furthermore, although the clinical complications of AS (HF, sudden death, atrial fibrillation) are mainly due to the consequences of AS on the LV, the impact of AS on LV function and morphology has rarely been studied in asymptomatic severe AS. Patients with concentric LV hypertrophy or remodelling, extensive LV fibrosis on cMRI or a marked decrease of LV strain by speckle tracking echocardiography may be at high risk of irreversible myocardial damage and clinical events. Baseline blood levels of cardiac biomarkers, such as B-type natriuretic peptide, ST2, high-sensitive troponin T and their changes over time may also be of interest. Hence, in addition to randomised trials, further large multicentre prospective observational studies with careful follow-up, extensive multimodality imaging and laboratory assessment of patients with asymptomatic AS are needed to identify the most useful parameters to identify high-risk subgroups of patients with true ‘very severe’ asymptomatic AS in terms of prognosis. While waiting for the results of these studies, a tailor-made management by a heart team in specialised heart valve clinics is recommended in patients with asymptomatic severe AS, with thorough risk stratification, patient education and close follow-up.


View Abstract


  • Competing interests None declared.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles