Catrin Williams and Daniel Obaid
Hywel Dda University Health Board
Aortic stenosis (thickening and narrowing of the aortic valve) is the most common cardiac valvular pathology requiring clinical intervention. Its prevalence increases with age and left untreated, has significant morbidity and mortality.
Detecting undiagnosed aortic stenosis is a significant undertaking and current strategies focus on detecting severe symptomatic aortic stenosis to allow intervention. However aortic stenosis effects mortality across all spectrums of severity and many patients are asymptomatic. Early detection provides an opportunity for surveillance and optimum timing of treatment. Cardiac auscultation has been a traditional method to detect valvular heart disease, however, it relies on the availability of trained clinicians and studies have shown poor diagnostic accuracy when screening valvular heart disease.
Utilising artificial intelligence (AI) has shown promise in improving the accuracy of diagnosing aortic stenosis without the need for trained clinicians. Recent developments combining AI-based software with digital auscultation allow identification of pathological cardiac murmurs with high diagnostic accuracy. However, these preliminary studies have been carried out in hospital environments and it is not known if it would be feasible in a community setting. Our study therefore aims to evaluate whether a digital stethoscope with AI algorithm can provide effective community screening for significant aortic valve stenosis.
Patients over the age of 65 years attending their local primary care practice will be offered ‘opportunistic screening’ for the detection of asymptomatic aortic stenosis. The screening will be undertaken by community practice nurses. Patients identified as having significant murmurs will be invited for echocardiography and the presence or absence of significant cardiac disease will be determined. Patients found to have significant valve disease will be placed on the appropriate clinical pathway. Patients determined to have valve disease requiring ongoing surveillance will be informed of the diagnosis and placed on to the valve surveillance clinic programme.
