June 12, 2022

expert reaction to a conference abstract on retinal screening predicting risk of myocardial infarction

A conference abstract presented at the European Society of Human Genetics Conference looks at the use of retinal vascular complexity to predict risk of myocardial infarction.

Professor Sir Nilesh Samani, Medical Director at the British Heart Foundation, said:

“The earlier and more accurately we can predict someone’s risk of heart attacks, the better the opportunities for prevention. This study suggests that information from retinal scans can improve such prediction. However, more research is needed to show that this improvement in prediction is robust. Work will also be required to understand the feasibility of this approach and determine how best to incorporate these scans into routine clinical practice.”

Dr James Ware, Cardiologist, Reader in Genomic Medicine at Imperial College London and MRC Investigator, MRC London Institute of Medical Sciences, said:

“The abstract is very concise and does not contain much methodological detail (as is often the case for a conference abstract). This doesn’t mean that there is any problem with the work, but I am not able to judge whether the study is robust from this short summary, and it has not been peer reviewed, so I would await more detail before forming an opinion on the robustness of the findings.

“There was a paper that looks very similar earlier this year in Nature Machine Intelligence (https://www.nature.com/articles/s42256-021-00427-7) also using imaging of retinal vessels to predict myocardial infarction (MI). There isn’t enough info in the abstract to make clear to me if this approach is importantly different or better. 

“And there is other prior work predicting coronary artery disease (though not specifically MI) https://www.thelancet.com/journals/landig/article/PIIS2589-7500(21)00043-1/fulltext, also using the UK biobank.

“It is well recognised that the retina provides a unique opportunity to directly visualise vessels and assess vascular health.  Approaches like this that use computer vision and/or machine learning to detect subtle vascular features predictive of future heart health appear promising.   

“This study also included genetic information in the prediction model – which would not normally be available at the time of a routine eye exam. It will be interesting to see whether the model identifies people at risk of MI without knowing their polygenic risk score, as this would be simpler to implement in practice. Genetic risk scores also promise to be very powerful tools for early identification of at-risk individuals, and indeed genetic risk can be assessed from birth.

“The studies also highlight the enormous value of the UK Biobank which is a hugely powerful research resource, and represents a fantastic investment in science in the UK, with far reaching benefits.”

The press release is based off an abstract ‘Decreased retinal vascular complexity is an early biomarker of myocardial infarction supported by a shared genetic control’ being presented at the European Society of Human Genetics Conference and was under embargo until 23:01 UK time Sunday 12 June.

Declared interests

Dr James Ware: “I don’t have any disclosures that are relevant.!

No others received.