Groundbreaking model predicts heart attack risk months in advance

NewsGuard 100/100 Score

In a recent study published in the journal Nature Cardiovascular Research, researchers developed and validated a novel biomarker-based prediction model to evaluate imminent first myocardial infarction. To train and test the model, they collected protein and metabolite data from a case-cohort consortium comprising 2,018 individuals from 6 preexisting European studies. Out of more than 800 analyzed proteins and 1,000 metabolites, 48 and 43 were found to be associated with short-term imminent myocardial infarction (IMI) risk. Their findings reveal that their model is capable of good discriminatory performance between patients at risk of IMI and those without, uses readily available clinical variables, and presents the best effort yet at preventing and preparing for IMI.

Letter: Markers of imminent myocardial infarction. Image Credit: Africa Studio / ShutterstockLetter: Markers of imminent myocardial infarction. Image Credit: Africa Studio / Shutterstock

Myocardial infarction and difficulties in its prediction

Myocardial infarction (MI) is a severe and often lethal cardiovascular condition caused by a decrease or complete cessation of blood flow to a portion of the myocardium. Colloquially known as a ‘heart attack,’ MI remains the leading cause of global non-transmittable death.

Despite decades of research that have focused on preventing MI, unfortunately, most risk prevention methods are unable to account for the dynamic nature of MI’s causative variables, significantly hampering efforts aimed at identifying high-risk imminent myocardial infarction (IMI) patients. Stochastic traumatic events such as the loss of a loved one or a cancer diagnosis are known to increase IMI risk substantially. Still, they are hard to predict, mainly because each person reacts to these variables differently.

A large population study aimed at identifying biomarkers of IMI is therefore essential in predicting and preparing for the condition. Biomarkers (circulating biomarkers) may provide reliable and quantifiable sources of data, and while they have been investigated in the past, most studies used small sample cohorts and limited follow-up periods, reducing reliability and confounding results. 

“Primary prevention for asymptomatic risk factors over a long period is costly, and motivation among patients and providers is limited even for secondary prevention. Risk prediction in the short term based on biomarkers of IMI might tilt the scales for prevention, as the knowledge of an increased risk of a first myocardial infarction within the ensuing few months might motivate patients and doctors to consider preventive strategies.”

About the study

The present study was conducted based on the hypothesis that circulating biomarkers may serve as proxies for critical yet dynamic biological processes that precede MI by months. This would provide patients and clinicians ample opportunity to implement mitigation strategies and prepare for IMI.

Data for the study was obtained from a nested case-cohort study entitled “The Markers of Imminent Myocardial Infarction (MIMI).” The study comprises biobanked blood (250 μl of plasma) and associated demographic and medical data from six European study populations belonging to the Biobanking and Biomolecular Research Infrastructure-Large Prospective Cohorts (BBMRI-LPC) collaboration. For this study, patients with a clinical history of cardiovascular disease were excluded, resulting in a collated sample size of 2,018 participants.

The outcome of interest was IMI (specifically, acute myocardial infarction) onset within six months of baseline blood collection. All included samples were subjected to protein and metabolite characterization using the Olink proximity extension assay and ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS), respectively.

For model training and testing, sample-specific protein, metabolite, and biomarker information were divided between discovery (70%) and validation (30%) datasets, randomized, and repeated 100 times to bolster effective sample size. IMI risk assessment was carried out using weighted, stratified Cox proportional-hazards regression models. While tested, least absolute shrinkage and selection operator (LASSO) regression models and random forest machine learning (ML) models were unable to identify biomarkers capable of improving risk prediction.

Study findings

Of the 2,018 individuals included in the study, 420 developed IMI within six months of baseline blood collection and were treated as cases, while the remaining 1,598 individuals were treated as subcohort representatives. Protein and metabolite analyses of preserved blood plasma samples revealed 817 proteins and 1,025 metabolites. When combined with demographic and clinical histories, 16 clinical variables were harmonized between cohorts.

Model training and subsequent validation revealed 48 proteins, 43 metabolites, and three clinical factors (variables: sex, age, and systolic blood pressure) associated with IMI risk. When adjusting regression models for age and sex, the brain natriuretic peptide (BNP) emerged as the strongest determinant of IMI risk.

“BNP was the only biomarker with a suggestive association in the internal validation, passing the formal replication criteria in 22 of 100 random splits. By comparison, stem cell factor (SCF) and interleukin-6 (IL-6), biomarkers with a weaker support of an association, replicated in only 5 or 4 of 100 random splits.”

To identify causative variables involved in dynamic biomarker levels, patients’ BNP values were compared against their coronary artery calcium score (CACS). The CACS test measures the extent of calcium deposits along the walls of the coronary artery, with higher values corresponding to higher IMI risk. Surprisingly, after correcting for demographic variables, no association between BNP and CACS was observed.

Encouragingly, the clinical risk prediction model developed during this study achieved an internally validated C-index of 0.78, highlighting its ability to discriminate between first IMI cases and low- to no-risk noncases. This score was further improved to 0.82 when using an external validation dataset from the United Kingdom (UK) Biobank.


“In the current study, higher BNP levels in individuals without a known cardiovascular disease were linked to a higher risk of a first myocardial infarction within 6 months in several models. Cardiomyocytes produce BNP in response to strain8, and NT-proBNP measurement is a pillar of the clinical management of heart failure but is not used in diagnosing myocardial infarction.”

The present study used the largest cohort to date to identify circulating biomarkers that could predict IMI in patients without a clinical history of cardiovascular disease. Their analyses of 2,018 plasma samples revealed 817 proteins and 1,025 metabolites, 48 and 43 of which were associated with IMI. Cox regression models adjusted for age and sex revealed BNP to be the best biomarker for IMI prediction.

Encouragingly, a novel algorithm trained using the above data was able to discriminate between IMI cases and noncases with a C-index of 0.78 to 0.82, making it the best predictive IMI model to date, and the ideal basis for motivating patients and clinicians towards primary prevention in at-risk individuals.

Journal reference:
Hugo Francisco de Souza

Written by

Hugo Francisco de Souza

Hugo Francisco de Souza is a scientific writer based in Bangalore, Karnataka, India. His academic passions lie in biogeography, evolutionary biology, and herpetology. He is currently pursuing his Ph.D. from the Centre for Ecological Sciences, Indian Institute of Science, where he studies the origins, dispersal, and speciation of wetland-associated snakes. Hugo has received, amongst others, the DST-INSPIRE fellowship for his doctoral research and the Gold Medal from Pondicherry University for academic excellence during his Masters. His research has been published in high-impact peer-reviewed journals, including PLOS Neglected Tropical Diseases and Systematic Biology. When not working or writing, Hugo can be found consuming copious amounts of anime and manga, composing and making music with his bass guitar, shredding trails on his MTB, playing video games (he prefers the term ‘gaming’), or tinkering with all things tech.


Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Francisco de Souza, Hugo. (2024, February 14). Groundbreaking model predicts heart attack risk months in advance. News-Medical. Retrieved on April 17, 2024 from

  • MLA

    Francisco de Souza, Hugo. "Groundbreaking model predicts heart attack risk months in advance". News-Medical. 17 April 2024. <>.

  • Chicago

    Francisco de Souza, Hugo. "Groundbreaking model predicts heart attack risk months in advance". News-Medical. (accessed April 17, 2024).

  • Harvard

    Francisco de Souza, Hugo. 2024. Groundbreaking model predicts heart attack risk months in advance. News-Medical, viewed 17 April 2024,


The opinions expressed here are the views of the writer and do not necessarily reflect the views and opinions of News Medical.
Post a new comment

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.

You might also like...
SGLT2 inhibitors: A game-changer in preventing heart failure and sudden cardiac deaths