I
ntroduction:
A groundbreaking AI tool, capable of predicting the 10-year risk of a heart attack, is poised to revolutionise the treatment landscape for patients undergoing CT scans to investigate chest pain. This transformative technology was unveiled in recent research presented at the American Heart Association’s Scientific Sessions in Philadelphia.
Real-world Impact:
In the first real-world trial, the AI tool exhibited a remarkable potential to enhance treatment for up to 45% of patients. Moreover, it holds the promise of potentially saving the lives of thousands suffering from chest pain, who may not have been previously identified as being at risk of a heart attack.
The Challenge of Cardiac CT Scans:
Annually, approximately 350,000 individuals in the UK undergo cardiac CT scans, the standard test for identifying narrowings or blockages in the coronary arteries. Sadly, in about three-quarters of cases, no clear signs of significant narrowings emerge, leading to patients being reassured and discharged. Regrettably, a considerable number of these individuals may later experience a heart attack due to small, undetectable narrowings that may break up if inflamed, causing artery blockages. Until recently, identifying these at-risk patients was a formidable challenge.
Harnessing the Power of Artificial Intelligence:
Led by Professor Charalambos Antoniades at the University of Oxford, a new study analysed data from over 40,000 people undergoing routine cardiac CT scans at eight UK hospitals. Over a median follow-up period of 2.7 years, the research revealed that while individuals with significant coronary artery narrowings were more prone to serious cardiac events or death, twice as many patients with no significant narrowings experienced heart attacks and cardiac deaths.
The AI tool, trained using information on changes in the fat around inflamed arteries, as well as details on artery narrowings and other clinical risk factors, showed remarkable potential. Extended testing confirmed its ability to independently and accurately predict the risk of cardiac events. Among those with unobstructed arteries, individuals with the highest levels of inflammation in their blood vessels faced more than a 10-fold higher risk of cardiac death compared to those with lower inflammation levels.
World-first Pilot:
In an unprecedented pilot initiative, the research team provided AI-generated risk scores to clinicians for 744 patients. In up to 45% of cases, clinicians altered patients’ treatment plans, underscoring the significant value of this AI tool in guiding and informing the management of patients with chest pain. This ensures early identification and preventative treatment for those at the highest risk.
Cost-effectiveness for the NHS:
Analysis comparing the use of the AI tool to standard care demonstrated its high cost-effectiveness for the NHS. Furthermore, researchers estimate that implementing this technology in the NHS could result in over 20% fewer heart attacks and 8% fewer cardiac deaths and strokes among those undergoing the test. With the technology required to power the AI tool already commissioned by NHS England for a pilot programme in five NHS hospitals, the researchers are optimistic about its potential widespread adoption across the UK.
Hope for the Future:
Professor Charalambos Antoniades, BHF Professor of Cardiovascular Medicine at the University of Oxford, expressed optimism about the future implementation of the AI tool across the NHS. He highlighted the potential to prevent thousands of avoidable deaths from heart attacks annually in the UK.
Medical Director’s Perspective:
Professor Sir Nilesh Samani, Medical Director, emphasised the valuable role AI-based technology can play in better identifying patients most at risk of future heart attacks. He stressed the need to harness AI’s potential to guide patient treatment, ensuring that the NHS is well-equipped to support its use and ultimately saving lives.
Research Support:
Professor Antoniades’s research is also supported by the National Institute for Health and Care Research Oxford Biomedical Research Centre.