Surgery in asymptomatic aortic stenosis

Aortic stenosis is the commonest valve disease in the developed world that requires surgery. In elderly people its prevalence is approximately 3%.¹ Usually a long latent period precedes the cardinal symptoms of the disease.² Replacement of the aortic valve is an excellent treatment for severe symptomatic aortic stenosis—symptoms improve in most patients and life expectancy approaches that of age matched controls.³ The risks of surgery are declining steadily. In patients who are having their first operation with a mechanical prosthesis and do not require a bypass graft, perioperative mortality is only 2%.⁴ No randomised controlled trials of aortic valve replacement in symptomatic aortic stenosis have been conducted, yet the level of evidence is regarded as class I because of the dramatic improvement in prognosis with surgery.⁵

In severe asymptomatic aortic stenosis the risks of perioperative morbidity, mortality, and long term complications related to the prosthesis have to be weighed against the risk of sudden cardiac death and the morbidity and mortality experienced on lengthy surgical waiting lists.⁶ The ideal time to operate is immediately before symptoms develop.

Early studies reported a low incidence of sudden death in asymptomatic aortic stenosis, but this was in the era before the severity of the stenosis could be quantified accurately.² Studies in the modern era have used echocardiographic measures of severity, and outcomes have been based on the combined end point of death or the need for replacement of the aortic valve. Increased peak aortic velocity (more than 400 cm/s), a fast rate of increase in peak aortic velocity (more than 30 cm/s per year), calcification of the valve, and increased age of the patient are predictors of outcome.⁷,⁸ In these reports only one sudden death occurred before symptoms developed, although another patient died waiting for surgery. The authors concluded that surgery should not generally be recommended in asymptomatic aortic stenosis and that patients could be monitored safely, consistent with guidelines published previously.⁵

Although prospective in design these studies have methodological flaws. In one report 50% (24) of the 48 patients undergoing surgery were asymptomatic at the time of operation (in 18 the decision was based solely on reduced exercise tolerance),⁷ which limits the conclusions. In the other report 22 of 128 asymptomatic patients were excluded from analysis because of early operative intervention.⁸ Excluding this high risk subgroup (older age and increased peak aortic velocity) probably reduced the overall event rate for death and development of symptoms of the study. In both studies doctors were not blinded to the results of exercise tolerance tests or echocardiography and therefore, almost certainly, influenced decisions to recommend surgery. The European heart survey on valve disease also shows that many cardiologists refer asymptomatic patients with aortic stenosis for surgery.⁹

When a rigid policy of not operating on asymptomatic patients is adhered to, the true incidence of sudden death approaches 5% per year.¹⁰ In this study of 66 patients with severe aortic stenosis, coronary artery disease and other serious comorbid conditions were excluded. Eighteen patients had their aortic valve replaced, and four died during follow up, which had a mean duration of 15 months. Exercise testing was the best predictor of outcome—a positive test had a high sensitivity of 92% but a relatively low specificity of 68%.¹⁰ The safety of exercise testing in aortic stenosis with careful haemodynamic and electrocardiographic monitoring is well established.⁷

The European Society of Cardiology has published guidelines on the management of asymptomatic valvular heart disease, taking into account these recent publications.¹¹ Owing to a lack of randomised controlled trials, the society thought that the strength of evidence supporting each specific recommendation could not be stated. Asymptomatic patients have been defined as those that can achieve 80% predicted maximum heart rate without developing symptoms. Surgery is recommended for severe aortic stenosis (area less than 1.0 cm² or 0.6 cm².m^-2 (aortic valve area corrected for body surface area) in the following situations: firstly, an abnormal response to exercise—if symptoms develop, if systolic blood pressure falls or a blunted systolic blood pressure response (less than 20 mm Hg) occurs when exercise capacity is poor; secondly, moderate to severe calcification of the valve, peak aortic velocity more than 400 cm/s, and a rate of progression more than 30 cm/s per year; thirdly, impaired left ventricular systolic function (ejection fraction less than 50%). Replacement of the aortic valve is also recommended in severe disease with the presence of marked left ventricular hypertrophy (more than 15 mm) or ventricular arrhythmias, although less evidence supports these strategies. Patients not fulfilling the above criteria should have follow up with echocardiography and exercise testing at intervals of 6-12 months, depending on the severity of disease.

These recommendations are a consensus statement rather than strong evidence based guidelines. The recommendations are likely to be refined as further studies of exercise testing and possibly natriuretic peptides are published.¹² In the meantime all asymptomatic patients with severe aortic stenosis should have an exercise test as a routine assessment and those with an abnormal response should be considered for replacement of the aortic valve. We, like most cardiologists, no longer believe that surgery is the most common cause of sudden death in asymptomatic patients with aortic stenosis.¹³