Calcium Scanning May Lower CAD Risk at No Extra Cost

Among asymptomatic subjects, adding coronary artery calcium (CAC) scanning to standard coronary artery disease (CAD) risk counseling results in long-term reduction in overall CAD risk without increasing downstream testing or management costs, according to a study published online March 23, 2011, ahead of print in the Journal of the American College of Cardiology.

For the EISNER (Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research) study, investigators led by Daniel S. Berman, MD, of Cedars-Sinai Medical Center (Los Angeles, CA), randomized 2,137 asymptomatic volunteers—primarily middle-aged individuals with CAD risk factors—in a 2:1 ratio to CAC scanning (n = 1,424) or no scan (n = 713) between May 2001 and May 2005.

At a baseline examination, a nurse practitioner provided standard risk counseling plus a customized packet containing American Heart Association guidelines on cardiac risk factors, subjects’ individual risk results, and information on how to improve risk profiles. With scanned individuals, the nurse practitioner also reviewed the CAC images and scores, informing them that the presence of any calcium was evidence of atherosclerosis. CAC test results were not sent to the subjects’ primary care physicians, although participants were encouraged to share the results with their doctors.

Impact on Multiple Risk Factors

At a 4-year follow-up visit, 88.2% of the scan group and 81.9% of the no-scan group (P < 0.001) underwent a repeat of assessments obtained at baseline and all underwent CAC scanning. Compared with the no-scan arm, those who were scanned at the outset had a greater reduction in systolic blood pressure and serum LDL cholesterol as well as a smaller waist circumference if they had a wide girth at baseline. Scanned subjects also tended to gain less weight if they were overweight at baseline (table 1).

Table 1. Median Changes in CAD Clinical Risk Factors at 4-Year Follow-up



No Scan


P Value

Systolic BP, mm Hg




LDL Cholesterol, mg/dL




Waist Circumference, in




Weight, lbs




Abbreviations: BP, blood pressure; LDL, low-density lipoprotein; HDL, high-density lipoprotein.

On the other hand, there was no difference between the groups with respect to HDL cholesterol, triglyceride, and glucose levels; smoking cessation; or new exercise activity.

Importantly, the mean Framingham Risk Score increased over baseline in the no-scan arm (0.7 ± 5.1) while it remained essentially unchanged in the scan arm (0.002 ± 4.9; P = 0.003).

In addition, more patients who underwent CAC scanning had initiated new antihypertensive medications (24% vs. 18%; P = 0.02) than those who were not scanned. There also was a trend toward increased use of lipid-lowering agents in the scanned vs. unscanned group (29% vs. 25%; P = 0.06).

Scanning Did Not Trigger More Testing

The 2 groups did not differ in the use of procedures such as stress tests, carotid ultrasound studies, noninvasive or invasive coronary angiograms, or revascularization procedures. Moreover, overall costs for medical procedures were comparable for both arms, but medication costs tended to be higher in the scanned group ($3,149 vs. $2,937; P = 0.09).

In the study population, there were 21 deaths (3 cardiac) and 12 MIs. The combined number of deaths/MIs was 1.0% in the unscanned group and 2.1% in the scanned group (P = 0.08).

Within the scan group, increasing baseline score was linked to proportionally greater improvement in most CAD risk factors at follow-up. This inverse relationship was seen for systolic and diastolic blood pressure, serum cholesterol, LDL cholesterol, and triglycerides (P for trend < 0.001 for all values). In addition, greater weight loss was observed among overweight subjects with CAC scores ≥ 100 at baseline, while for those with large waistlines, the greater decline occurred among those with CAC scores ≥ 400 at baseline. There also was a trend toward more exercise with increasing CAC scores. Framingham Risk Score rose in subjects with a CAC score of 0 but decreased in those with evidence of CAC at baseline.

The same inverse relationship with CAC score was observed for new cardiac medications, especially lipid-lowering agents. The frequency of procedures and procedural costs also increased with increasing baseline CAC scores. Procedural costs were low for all participants with no CAC, but much higher for those with CAC scores ≥ 400.

At 4 years, there was no difference between the no-scan group and the subset of scanned subjects with normal CAC scores (about half) in terms of adherence to baseline medications or CAD risk profile. However, the normal CAC subjects used fewer downstream tests or invasive procedures, incurring 37% lower procedure costs (P = 0.001) and 26% lower medication costs (P = 0.005).

At 4 years, mean CAC scores were comparable for the scan and no-scan groups (median 12 vs. median 11; P = 0.89), with similar distributions of scores. Interestingly, in a treatment-plus-baseline model, the strongest predictor of CAC score progression was use of a lipid-lowering medication at year 4.

Clinical Importance Remains to Be Seen

“Further work should similarly assess patients who are suitable candidates for CAC scanning based on clinical consensus and current guidelines,” the authors write.

Future large-scale trials are needed to determine whether the findings apply to different patient populations and whether the positive effect of CAC scanning on CAD risk profiles translates into improved long-term outcomes, the investigators say.

“I’m not so sure the data are generalizable across the population,” Michael Poon, MD, of Stony Brook University Medical Center (Stony Brook, NY), told TCTMD in a telephone interview. “The study seems to have enrolled a lot of sophisticated and affluent people. [After getting their calcium scores,] they were likely to be in the gym every day,” he said.

Another unrealistic aspect is that the subjects’ primary care physicians (PCP) were not necessarily informed of the test results, Dr. Poon indicated. “If a PCP does a calcium test on a patient and it comes back 450, he’s going to say, ‘Gee, I have a treadmill in my office. I should put him on the treadmill to see if he’s truly asymptomatic,’” he said. “So if this were done in the real world, I think it probably would lead to more downstream testing, which would add cost to the whole process.”

CAC Score Gets Patients’ Attention

Nonetheless, Dr. Poon found the main finding of a reduction in CAD risk after CAC scanning plausible. Motivation is the key, he said.

“I do calcium scoring on asymptomatic patients with a couple of risk factors when I want to impress on them that their coronaries are not normal,” Dr. Poon observed. Seeing a scan gets their attention in a way that cholesterol numbers do not, he indicated. “Most people look at cholesterol the way they look at the weather—it’s goes up a little, it comes down a little. But you show them white plaque right in the middle of the heart and tell them it can cause a heart attack, and it scares them.”

These findings may encourage more liberal use of calcium scanning among clinicians, Dr. Poon said, although it entails a lot of work for little compensation. He lamented the fact that insurance companies do not pay for the test in asymptomatic patients “because they consider it screening and to them screening means ‘fishing.’”

But the investigators are tying to change that perception by underlining “the intangible element of human psychology in patient care,” Dr. Poon suggested, adding, “The take-home message is that calcium score affects patients’ behavior.”

One drawback, however, is that calcification is usually irreversible and may progress despite aggressive medical management. In that regard, “calcium score doesn’t look so good, because it doesn’t help track improvement,” Dr. Poon admitted.

Study Details

The 2 groups were comparable in terms of age, sex, socioeconomic factors, cardiac risk factors, medication use, and Framingham Risk Score.

Scanning was performed using electron beam (GE Medical/Imatron, San Francisco, CA) or multislice CT (Siemens Medical Systems, Forchheim, Germany).

  • Rozanski A, Gransar H, Shaw LJ, et al. Impact of coronary artery calcium scanning on coronary risk factors and downstream testing: The EISNER (Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research) prospective randomized trial. J Am Coll Cardiol. 2011;Epub ahead of print.

  • Dr. Berman reports receiving research grants from Astellas GE/Amersham, Lantheus, and Siemens and serving on the speaker’s bureaus for Astellas and Lantheus.
  • Dr. Poon reports serving on the speaker’s bureau for Toshiba.