GPs in London and Wales will be given AI stethoscopes to help them diagnose heart failure as part of a new trial.
Funded by £1.2m from the National Institute of Health and Care Research (NIHR), the study will provide 100 GP practices with AI-enabled smart tools which can quickly show whether the pumping action of the heart is weakened.
Previous studies assessing the effectiveness of the technology – named the Eko DUO device – showed 91% sensitivity and 80% specificity, and that it was able to provide an immediate diagnosis of heart failure.
Heart failure affects 2% of the UK population and takes up 4% of the NHS budget, according to the researchers.
Currently, diagnosis for heart failure begins with the NT-proBNP blood test, but this may need to be followed by other tests including an ECG, an echocardiogram, or a chest X-ray.
The researchers argue that the ‘widespread use of AI tools in primary care has the potential to improve heart disease outcomes for millions of people in the UK’.
The trial will seek to determine this, by comparing outcomes from 100 GP practices who use the AI stethoscopes with outcomes from 100 GP practices in the same areas who continue with usual care for their patients.
GP practices using the technology will be able to give patients a brief, non-invasive heart exam using the stethoscope and accompanying AI app. If this shows signs of possible cardiac disease, GPs can quickly begin further tests and treatments.
Professor Nicholas Peters, of the National Heart and Lung Institute at Imperial College London, highlighted that heart failure costs the UK over £2bn annually and that ‘an unacceptable 80%’ of diagnoses are made during emergency admissions.
Co-investigators Dr Patrik Bächtiger and Dr Mihir Kelshiker, also from Imperial’s Heart and Lung Institute, said: ‘Implementing this tool in primary care could save the health payer system £2,400 per patient, eliminating the potential need for an emergency room visit.
‘Currently, four out of five patients are diagnosed too late, after emergency hospital admission, and they are likely to die prematurely.
‘GPs have no accurate, easy-to-use tools to quickly check for heart failure. And if they suspect a patient may have heart failure, patients may go through a long and rarely completed pathway to diagnosis.’
The study will take place at 200 practices across North West London and North Wales, and is coordinated by the TRICORDER programme – a collaboration between the North West London ICB, Betsi Cadwaladr University Health Board, and Imperial College London.
Last week, ICSs were asked by NHS England to expand their use of virtual wards to include patients with heart failure, meaning specialists – which includes GPs with a special interest in heart failure – should discuss this possibility with patients.
And last month, NICE recommended empagliflozin as a treatment option for some patients with chronic heart failure, making it the second NICE-approved treatment for the condition.
Noctors all the way. No need for a GP. Just press play.
‘80% of diagnosis of HF made during emergency admissions’, yes that was the case in 2010-2013 (Bottle et al, 2017). But isn’t that why we’ve been doing all those pro-BNP tests and echo’s for the last 10 years? More GP bashing.
(DOI I’ve seen a lot of cardiologists in the past 12 months and none of them has ever used a stethoscope to listen to my chest, AI or otherwise, not once. They use an echo and sometimes even stick it down your gullet)
Where can I buy one (before retraining as a PA) and then donate redundant Littmann Cardiology tubes to a museum?