Every year, hundreds of thousands of people suffer a heart attack without anything in their medical history having predicted it. However, cardiovascular risk assessment tools such as ASCVD or PREVENT scores are systematically used in preventive medicine. A rigorous study conducted by researchers at the Icahn School of Medicine at Mount Sinai, published in JACC: Advances, directly calls into question their effectiveness.
By analyzing data from 474 patients aged under 66 and hospitalized for a first heart attack, the authors reveal that almost half would have been considered non-priority for preventive treatment, according to the usual criteria. This observation highlights a structural flaw in the current approach: these scores, designed for the general population, fail to detect individuals truly at risk. This observation calls for a fundamental rethinking of cardiac prevention.
Statistical tools that miss their target
The ASCVD (Atherosclerotic Cardiovascular Disease) and PREVENT scores are today the pillars of cardiovascular risk assessment in clinical practice. Designed to estimate the risk of having a heart attack or stroke over 10 years, these tools are based on well-known variables: age, sex, ethnic origin, cholesterol, blood pressure, smoking or diabetes. The objective is to identify people at high risk in order to offer them drug prevention, particularly statins, or enhanced monitoring. But their ability to predict heart attacks individually seems largely overestimated.
The study carried out by Mount Sinai researchers highlights a major weakness. 45% of patients who had a heart attack would have been classified as “low” or “borderline” risk two days before the event, according to the ASCVD score. This percentage rises to 61% with the PREVENT score, although designed to be more sensitive. This observation does not arise from a specific defect, but from a structural bias. These tools were calibrated on population data and not on individual trajectories.
In other words, they remain effective at scale. But they fail to detect high-risk cases that do not present the classic profiles. These scores systematically underestimate the danger in young, active people without symptoms, but in whom silent atherosclerosis can develop without warning. Dr. Ahmadi emphasizes that “ the current approach, which relies on these scores as main filters, is no longer suitable for individualized medicine “. The problem concerns less the tool itself than its unique reference status. It can induce a false sense of security.
Symptoms too late to prevent effectively
One of the most striking results of the study is based on the time of onset of symptoms in patients. Of the 474 cases studied, 60% of individuals only experienced clinical signs in the 48 hours preceding their infarction. This extremely short period makes any symptomatic prevention ineffective. In other words, by the time symptoms appear, myocardial ischemia is already well underway, even irreversible.
This data calls into question a second pillar of current prevention. Namely: waiting or detecting clinical signs as a trigger for an assessment. In practice, a patient without chest pain or difficulty breathing, and displaying a low ASCVD score, will not benefit from any additional examination or preventive treatment. His doctor reassured him, as is standard practice. However, the study shows that these silent profiles constitute a significant proportion of victims of a first heart attack.
This situation is explained by the very nature of atherosclerosis. Plaques forming in the coronary arteries can remain stable and asymptomatic for years. But when they rupture, they trigger a blood clot which can suddenly block the artery. No pain precedes this mechanism, in the majority of low or intermediate risk cases.
Anna Mueller specifies: “ The absence of symptoms is not a guarantee. Many patients have experienced their first heart attack without warning signs “. In summary, relying on clinical signs to anticipate a cardiovascular event often means intervening too late. The disease appears to be already underway, silent, but active. It then leaves little room for maneuver once it becomes apparent.
A rigorous methodology for an alarming observation
The study is based on a precise methodology: a retrospective analysis of data from 474 patients, all aged under 66, with no known history of coronary heart disease, hospitalized between January 2020 and July 2025 for a first heart attack. These patients were treated at two Mount Sinai Health System hospitals in New York: Mount Sinai Morningside and The Mount Sinai Hospital.
The researchers collected for each: demographic data, medical history, cholesterol level, blood pressure, smoking status, diabetes, and date of onset of first symptoms. Then, they retroactively calculated the ASCVD score on D-2 (two days before the infarction). As if these patients had been evaluated in a routine consultation.
The results are clear. 209 patients (45%) would not have been referred for preventive care according to recommendations based on ASCVD. With the PREVENT score, this figure rises to 285 patients (61%). These data cover a cohort of patients followed in a real clinical context, outside of therapeutic trials, far from anecdotal data.
The researchers also categorized the scores into four levels: low (less than 5% risk at 10 years), borderline (5 to 7.5%), intermediate (7.5 to 20%), and high (more than 20%). This standardized classification makes it possible to better interpret the clinical thresholds used to decide whether or not to treat with statins or to carry out additional testing. But here, it reveals that the majority of heart attacks affect profiles not considered priorities. This gap between estimated risk and actual risk is at the heart of the problem. It highlights the limits of an approach based exclusively on population models.
Towards an overhaul of cardiovascular prevention strategies
Faced with these findings, the authors of the study call for a questioning of current paradigms. The prevention of cardiovascular diseases can no longer be satisfied with mathematical scores and the observation of symptoms. Amir Ahmadi suggests putting cardiovascular imaging back at the center of screening, even in patients without symptoms and a low score. The objective: to detect silent atherosclerotic plaques early, before they rupture.
Several non-invasive techniques already exist. Ultrasound of the carotid arteries, coronary CT or even the calcium score make it possible to visualize plaque deposits and assess the degree of arterial obstruction. These examinations, still little used in routine medicine, would target more precisely individuals truly at risk.
The challenge remains economic and logistical. Generalizing imaging for all patients at low or intermediate risk would raise questions of cost, availability, and clinical relevance. But the results of the study show that as they stand, current tools miss a significant portion of preventable cases. As Dr. Mueller summarizes, “ we must move from reactive medicine to proactive medicine. Medicine focused on detecting disease before it manifests “.
This approach is in the direction of personalized medicine, integrating more clinical, biological and imaging parameters. The future could also rely on artificial intelligence to cross-reference heterogeneous data and build dynamic risk profiles. In the meantime, healthcare professionals are urged not to rely solely on risk scores and to consider a more detailed assessment for certain patients, even without an obvious risk factor.
Source: Mueller, A, Leipsic, J, Tomey, M. et al. “Limitations of Risk- and Symptom-Based Screening in Predicting First Myocardial Infarction”. JACC Adv. 2025 Dec, 4
With an unwavering passion for local news, Christopher leads our editorial team with integrity and dedication. With over 20 years’ experience, he is the backbone of Wouldsayso, ensuring that we stay true to our mission to inform.
