Calcium Score Better Than Age for Discriminating Coronary Risk in Older Adults
These data bolster the existing advice to consider calcium scoring in risk assessment and may eventually change guidelines, one expert says.
Measuring coronary artery calcium (CAC) is superior to considering age when it comes to distinguishing older adults with an elevated risk of atherosclerotic cardiovascular events from those who carry a low risk, a finding that may have implications for clinicians trying to decide which patients to start on statins.
When added to a model that included traditional cardiovascular risk factors but left off age, CAC score improved prediction of incident coronary heart disease compared with a model that included age (C statistic 0.735 vs 0.703), according to lead author Yuichiro Yano, MD, PhD (Northwestern University Feinberg School of Medicine, Chicago, IL), and colleagues.
CAC score also was better than age for improving risk reclassification for incident cardiovascular events, they report in a study published online July 26, 2017, ahead of print in JAMA Cardiology.
I think it’s a potentially practice-changing observation. Philip Greenland
“We thought this was a pretty exciting paper actually,” senior author Philip Greenland, MD (Northwestern University, Chicago, IL), told TCTMD.
Greenland was a member of the expert group that worked on the 2013 American College of Cardiology/American Heart Association guideline on cardiovascular risk assessment, part of a larger prevention package. That document calls for use of the Pooled Cohort Equations to assess 10-year risk of a first hard atherosclerotic CVD event. If there is remaining uncertainty about the best course of action after that, the authors recommend using one of four additional factors—family history, high-sensitivity C-reactive protein, ankle-brachial index, or CAC score—to help make the decision.
“My prediction would be that if the guidelines were going to be rewritten today, there might be a stronger recommendation for coronary calcium [scoring] based on data like this,” Greenland said, referring to the superior predictive ability of CAC scoring versus age. “So I think it’s a potentially practice-changing observation.”
The 2013 cholesterol guidelines include a strong recommendation to initiate statin therapy in patients with a 10-year cardiovascular risk of 7.5% or higher after a discussion between the patient and his or her doctor. Greenland pointed out that using that threshold, most patients 60 and older would be considered candidates for statin therapy based on age alone. Thus, alternative approaches to risk prediction beyond age are warranted.
Low Risk With CAC Score of 0
To explore the potential of CAC scoring, the investigators pooled data from three US population-based studies—the Framingham Heart Study, the Multi-Ethnic Study of Atherosclerosis, and the Cardiovascular Health Study. They then validated the findings by pooling two European cohorts from the Rotterdam Study and the Heinz Nixdorf Recall Study. The analyses included individuals who were 60 or older and had no known atherosclerotic cardiovascular disease at baseline and who underwent CAC scoring with cardiac CT between 1998 and 2006.
The US studies included 4,778 participants followed for a median of 10.5 years. At baseline, 31% had a CAC score of 0.
During follow-up, total atherosclerotic cardiovascular events, coronary heart disease events, and stroke occurred in 11%, 8%, and 16% of patients who had no coronary calcium at baseline. Corresponding figures for those who had a CAC score of at least 300 were 42%, 45%, and 38%. The probability of remaining free from a cardiovascular event during 12 years of follow-up remained above 90% for patients with a CAC score of 0.
CAC score was more strongly associated with incident coronary heart disease than was age (C statistic 0.733 vs 0.690), with a more modest difference for stroke (C statistic 0.695 vs 0.670).
The addition of CAC score to predictive models generally improved discrimination for incident coronary heart disease, but not for stroke.
Similar findings came from analysis of the European cohorts.
Enough to Change Guidelines?
These findings demonstrate that CAC score may be better than age for discriminating risk of coronary disease, which could come into play when deciding on statin initiation, but it would still probably be helpful to confirm the observation in a clinical trial, Greenland said.
“For the time being, this is the best evidence that we have that if you’re trying to make decisions in older people, rather than just relying on age alone and the risk prediction that you would get from the standard calculators, coronary calcium would be a useful way of narrowing down and potentially eliminating some patients from the consideration of statins,” he said. For example, patients with a CAC score of 0 might not need to start taking the lipid-lowering drugs.
Commenting for TCTMD, Tasneem Naqvi, MD (Mayo Clinic, Phoenix, AZ), agreed with Greenland in saying that the results of this pooled analysis are potentially practice-changing, but she cited some caveats.
This is very solid data, and it may actually bring about some changes in our guidelines. Tasneem Naqvi
The authors make no mention of incidental findings on the cardiac CT scans—such as malignancies, imaging artifacts, and other abnormalities—and what impact that had on the resources used to follow up on and manage them, Naqvi said. She also pointed out that CAC scoring was not very good at discriminating risk of stroke.
In addition, Naqvi was hesitant to endorse avoiding statin therapy in patients with no coronary calcium. Risk of total atherosclerotic cardiovascular events was low but not nonexistent in that group during follow-up (11% overall or roughly 1% per year).
“I feel that 10% is a high risk still because the cutoff for risk by definition is 7.5%,” she said. “But at least you can discuss this risk with the patient and they may say, ‘You know what, I’m willing to take that 1% per year risk of having a heart attack rather than taking statins’ or ‘I’ve tried statins and it really bothers me, and I would rather live with that risk.’ So it certainly clarifies the scenario much more for these people based on their calcium score.”
That said, these data likely could be used to elevate CAC scoring over the other three recommended supplementary tests in the risk assessment guidelines, Naqvi indicated.
The findings “probably put calcium scoring as the only test to consider when you are not sure about what decision to make, because the other tests have really not been shown to be as powerful as calcium scoring has in the prior studies,” she said. “This is very solid data, and it may actually bring about some changes in our guidelines.”
Yano Y, O’Donnell CJ, Kuller L, et al. Association of coronary artery calcium score vs age with cardiovascular risk in older adults: an analysis of pooled population-based studies. JAMA Cardiol. 2017;Epub ahead of print.
- Yano, Greenland, and Naqvi report no relevant conflicts of interest.