Universal Risk Tool Works Well in Both Primary and Secondary Prevention

A risk calculator that incorporates established cardiovascular risk factors and additional cardiac biomarkers, such as high-sensitivity C-reactive protein (hs-CRP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP), accurately predicts the risk of future events in patients with and without established atherosclerotic cardiovascular disease (ASCVD), a new study shows.

Importantly, the universal risk assessment approach identified a significant proportion of individuals without preexisting ASCVD who had risks similar to or greater than what was seen in those with prior disease, say investigators.

“Using our equation, we observed that the top 20% of people without atherosclerotic cardiovascular disease actually have a higher risk than the bottom 40% of people with cardiovascular disease,” senior investigator Kunihiro Matsushita, MD, PhD (Johns Hopkins Bloomberg School of Public Health, Baltimore, MD), told TCTMD.

Video Topical Discussion

TCTMD's Quick Takes: Dr. Kunihiro Matsushita

March 6, 2024

Matsushita said that for patients the universal approach could be helpful in quantifying risk across a lifetime.

“We believe it can provide an advantage in terms of more continuous care from before they have an event to after they have one,” he said. “They are able to recognize a difference, for example, that before they developed cardiovascular disease their risk might be 5% but now it’s 15% after developing it. In that case, they may be more motivated to change their lifestyle or adhere to treatment.”

The universal approach is based on the awareness that there is substantial heterogeneity in risk among patients with and without ASCVD. For example, there are patients who have established ASCVD with excellent control of various risk factors, such as LDL cholesterol or blood pressure, while there are those without prior ASCVD who have an unhealthy lifestyle and/or hypercholesterolemia or hypertension.

At present, the American College of Cardiology and the American Heart Association (ACC/AHA) recommend the use of the pooled cohort equation (PCE) to identify primary-prevention patients who would benefit from lipid-lowering therapy or antihypertensive medication based on their 10-year risk of ASCVD. In secondary prevention, risk classification has largely focused on short-term prognosis in patients with preexisting ASCVD.

“The concept of primary and secondary prevention has been the cornerstone of guiding treatment,” said Matsushita. “In the past, this made sense because people with cardiovascular disease had a really high risk while those without it had a much lower risk.”

Matsushita pointed out that secondary-prevention therapies have improved dramatically in the last number of years and that it’s more important now to risk stratify patients with preexisting disease to help select those who benefit from more aggressive treatment. Even the new cholesterol guidelines make a distinction between high and very high risk to guide treatment decisions with some of the more potent and expensive lipid-lowering medications. 

“Depending on the risk, the direction of treatment could be different,” said Matsushita.

ASCVD Risk on a Spectrum

The new study, published online ahead of the February 6, 2024, issue of the Journal of the American College of Cardiology with Yejin Mok, PhD (Johns Hopkins Bloomberg School of Public Health), as lead author, explored the possibility of developing a long-term risk calculator for those with and without ASCVD. To do so, they evaluated whether established risk factors, as well as additional cardiac biomarkers, might be able to predict the risk of MI, stroke, and heart failure (HF) in the ARIC study cohort. To develop the model, they tested established risk factors from the PCE and considered various risk enhancers from the ACC/AHA cholesterol guidelines, as well as high-sensitivity cardiac troponin T (hs-cTnT), among others.

Over a median follow-up of 18.9 years, 3,209 participants (35%) developed MACE. In the base model, each risk factor was associated with a similar risk of MACE in those with and without ASCVD. In the LASSO regression analysis, the investigators identified 10 variables that universally predicted MACE in patients with and without ASCVD, including age, diabetes, systolic blood pressure, antihypertensive medication, smoking status, hs-CRP, NT-proBNP, and hs-cTnT. The model also included interactions between age and total cholesterol with history of ASCVD.  

Overall, the C-statistic for predicting MACE with the model was 0.747 for patients without ASCVD and 0.691 for those with preexisting disease. The model also had good calibration in both cohorts (calibration slope 1.05 and 0.92 in the no-ASCVD and ASCVD groups, respectively). The risk model was also validated in 5,322 patients enrolled in the MESA study.

“These C-statistics are similar to the existing risk prediction models that we are using,” said Matsushita.

For clinicians, one advantage is that the calculator will allow them to identify patients without ASCVD who have a future predicted risk akin to someone with a history of disease. This concept is not completely new, said Matsushita, noting that a diagnosis of diabetes is considered a risk equivalent that shifts patients into a higher-risk bucket even without an ASCVD diagnosis. In secondary prevention, the calculator gives doctors a better sense of longer-term risk prediction that isn’t present with existing calculators.

In an editorial, Pier Sergio Saba, MD, PhD (Sassari University Hospital, Italy), along with Sadeer Al-Kindi, MD, and Khurram Nasir, MD, MPH (both Houston Methodist DeBakey Heart & Vascular Center, Texas), write that a universal assessment of risk across the spectrum would be a beneficial contribution to cardiovascular care. Given the wide variation of risk within the primary- and secondary-prevention buckets, a binary, categorical classification system is too simplistic, they write, adding that this continuous-risk concept could allow doctors to intensify treatment in the absence of ASCVD.

Like Matsushita, the editorialists say the model needs to be validated in prospective cohort studies and real-world patients, and that it should be compared with existing models. Additionally, given the use of cardiac biomarkers, a careful cost-benefit analysis is necessary. Still, they believe that universal risk assessment could also have an impact in the design of future research trials, noting that investigators can also move away from binary ASCVD/no-ASCVD inclusion criteria.