Aortic stenosis (AS), or narrowing of the aortic valve, is the commonest condition requiring valve surgery in the developed world. It is currently not known what determines who will go on to develop symptoms. Exercise testing may be able to identify these patients better than the severity of the narrowing itself, but with some limitations.

The purpose of this study is to compare whether MRI scanning or exercise testing can better identify patients with AS who are likely to benefit from surgery.

Design: The investigators will measure blood flow to the heart muscle with MRI scanning and perform exercise testing in 170 patients with AS and follow them for up to up to 2 years. Expected outcomes: MRI scanning will more accurately identify those patients with AS who will need surgery during this period. Anticipated Health Benefits: improved selection of patients with AS who are likely to benefit from early surgery. This is likely to reduce deaths in such patients.

Surgical AVR remains the universally accepted management for symptomatic aortic stenosis (AS). However, the best management of severe aortic stenosis, in the absence of symptoms, remains one of the most controversial areas in modern Cardiology.

Exercise testing can identify asymptomatic patients with AS at increased risk, but with limited specificity. In a BHF funded project, the investigators have identified that cardiac MRI measured Myocardial Perfusion Reserve (MPR) may be a novel imaging biomarker in AS. MPR was the only independent predictor of aerobic exercise capacity (peak VO2) in patients with severe AS and was also inversely related to symptomatic status.

In this multi-centre, observational, cohort outcome study, the investigators will follow 175 patients with asymptomatic moderate to severe AS for a minimum of 12 months, and determine whether MPR is a better predictor of outcome than exercise testing, elucidate the mechanisms contributing to symptom development in AS and establish the determinants of MPR in AS. Patients will be recruited from tertiary Cardiac centres, as well as regional hospitals. Comprehensive CMR with adenosine stress to determine LV mass and function, focal and diffuse fibrosis and MPR; cardiopulmonary exercise testing (peak VO2 and exercise symptoms); rest and exercise echocardiography (AS severity, valve compliance) and NT-proBNP will be performed. The study will be run in conjunction with the Glasgow CTU. Investigations will be analysed blind to patient status and data will be entered in a validated database. Statistical analysis will be performed under the supervision of Prof. Ian Ford. The relationship between MPR and exercise testing with 1-year outcome will be analysed using logistic regression. Paired comparisons of the specificities of the two approaches on the same dataset will be carried out using McNemar's test.

The primary hypothesis is that MPR will be a better predictor of adverse outcome than exercise testing.

Inclusion Criteria:

1. Moderate-severe aortic stenosis (2 or more of: AVA < 1.5cm2, peak PG >36mmHg or mean PG > 25mmHg).
2. Asymptomatic.
3. Age > 18 years and < 85 years.
4. Prepared to consider AVR if symptoms develop.
5. Ability to perform bicycle exercise test

Exclusion Criteria:

1. History of CABG or MI within previous 6 months.
2. Severe valvular disease other than AS.
3. Previous Valve surgery
4. Persistent Atrial Fibrillation or Flutter
5. History of Heart Failure
6. Severe Asthma.
7. Severe renal impairment eGFR < 30ml/min.
8. Planned aortic valve replacement.
9. Significant LV systolic dysfunction (EF < 40%)
10. Any absolute contraindication to CMR
11. Any absolute contraindication to Adenosine
12. Participation in an Interventional Clinical Trial at Inclusion.
13. Other medical condition that limits life expectancy or precludes AVR.
14. Pregnancy

Cardiology outpatients department and echocardiography department.