Can Coronary Artery Calcium Be Used to Assess Long-Term Heart Failure Mortality Risk?

During the median 12.5-year follow-up, there were 260 heart failure-related deaths.

Most occurred among people with substantial coronary calcium. 75.3% of heart failure-related deaths occurred in participants with baseline CAC >100, and 138 of the 260 deaths occurred among participants with CAC ≥400.

Compared with participants with CAC = 0, the adjusted association with heart failure-related mortality increased progressively as CAC increased:

CAC 1–99: SHR 2.27; 95% CI 1.30–3.99; CAC 100–399: SHR 3.68; 95% CI 2.10–6.43; CAC ≥400: SHR 7.05; 95% CI 4.05–12.29

The overall trend across CAC categories was statistically significant.

Absolute event rates remained relatively low. Heart failure-related mortality was below 0.6 per 1 000 person-years for CAC below 400 and approximately 1.67 per 1 000 person-years for CAC ≥400.

The association remained after adjustment for age, sex, study site, hypertension, dyslipidemia, family history of coronary artery disease, diabetes and smoking.

The relationship was observed in both women and men and across age groups. However, the gradient associated with increasing CAC was steeper among women and among participants younger than 65 years.

For women, CAC ≥400 was associated with an adjusted SHR of 10.46 compared with CAC = 0. For men, the corresponding SHR was 5.05.

Among participants younger than 65 years, CAC ≥400 was associated with an SHR of 15.40, compared with 5.04 among participants older than 65 years.

The researchers also examined CAC within cardiovascular risk categories calculated using both PCE and PREVENT.

Higher CAC continued to distinguish heart failure-related mortality risk within these groups. In particular, CAC ≥400 identified substantially higher risk even among participants classified as low or borderline risk by conventional risk equations.

A sensitivity analysis excluding deaths in which the underlying cause was non-cardiovascular showed a similar stepwise relationship between increasing CAC and heart failure-related mortality.

A coronary calcium score is primarily a measure of calcified coronary atherosclerotic burden. It is not a measurement of heart function and it does not diagnose heart failure.

This study nevertheless found a strong graded association between the amount of coronary calcium already present and long-term heart failure-related mortality.

That distinction is important.

Heart failure develops through multiple pathways and may be associated with ischemic heart disease, hypertension, diabetes, obesity, valvular disease and other myocardial disorders. The study does not establish which mechanisms explain the association between CAC and later heart failure-related death.

Instead, CAC may act as a marker of accumulated cardiovascular disease and risk over time.

The population studied is particularly relevant to prevention: participants were asymptomatic adults without known coronary artery disease undergoing cardiovascular risk assessment.

The results therefore suggest that evidence of subclinical coronary atherosclerosis may contain prognostic information extending beyond traditional estimates of atherosclerotic cardiovascular risk.

The findings also illustrate why cardiovascular longevity cannot be assessed from a single risk factor or risk equation alone. Clinical risk factors and imaging markers describe different aspects of cardiovascular health.

CAC scoring and cardiovascular risk calculators provide complementary types of information.

Risk equations such as PREVENT estimate future cardiovascular risk from clinical variables including age, blood pressure, cholesterol, kidney function, diabetes and smoking.

A coronary calcium scan provides anatomical information: it demonstrates calcified coronary atherosclerosis that is already present.

The present study is particularly interesting because CAC continued to differentiate heart failure-related mortality within categories of estimated cardiovascular risk.

This was visible not only in people already classified as intermediate or high risk. Participants with apparently lower estimated risk but CAC ≥400 also had substantially higher heart failure-related mortality than those with CAC = 0.

This does not mean that CAC should replace cardiovascular risk calculators. Rather, the two approaches may provide different information that can contribute to a more individualized cardiovascular assessment.

The findings also require several important qualifications.

First, this was a retrospective observational cohort study, not a randomized clinical trial. It demonstrates an association between CAC burden and heart failure-related mortality but cannot establish that coronary calcium causes heart failure-related death.

Second, the study did not test whether performing CAC screening or changing treatment according to the CAC result reduces heart failure mortality.

Third, only 260 heart failure-related deaths occurred among 66 636 participants. The relative associations were substantial, particularly at high CAC levels, but the absolute event rates were low.

Fourth, heart failure-related mortality was identified from death certificates. The study assessed mortality related to heart failure, not the incidence of new heart failure, heart failure hospitalization, cardiac function or all-cause mortality.

Fifth, the population was predominantly White and consisted of people referred for CAC testing at four US centers. This limits how directly the findings can be generalized to other populations.

Finally, although statistical models adjusted for major cardiovascular risk factors, residual confounding cannot be excluded.

The authors concluded that people with high CAC may warrant greater preventive attention directed toward both heart failure and atherosclerotic cardiovascular disease, including lifestyle measures and appropriate pharmacotherapy. However, this recommendation should be distinguished from the study’s direct evidence: the study demonstrated an association; it did not test whether CAC-guided preventive treatment improves clinical outcomes.