Abstract & Commentary
Synopsis:A calcium score < 100 eliminates the need for MPS, but patients with a negative MPS often have coronary calcium. These findings imply a potential role for applying coronary artery calcium screening after MPS among patients manifesting normal MPS.
Source: Berman DS, et al. J Am Coll Cardiol. 2004;44: 923-930.
Coronary artery calcium detected by fast x-ray computed tomography (CT) or electron beam CT is associated with the presence of coronary atherosclerosis. Stress myocardial perfusion scans (MPS) can identify intermediate-risk patients likely to have coronary artery disease (CAD). The potential predictive relationship between coronary calcium and the likelihood of stress-induced myocardial ischemia is not well understood. Thus, Berman and colleagues studied 1195 patients referred for MPS on clinical grounds who also had CT for coronary calcium done. Exercise stress was used in 88%; the remainder had adenosine. Rest thallium-201 images were compared to post stress technetium-99 images. MPS and CT were done within 7 days on average. No patient had known CAD, and 51% were asymptomatic.
Results: The 76 patients with a stress MPS positive for ischemia exhibited more CAD risk factors, more abnormalities on stress testing, and higher calcium scores compared to those with a normal MPS. There was a crude relationship between calcium score and a positive MPS: calcium score < 100, positive MPS < 2%; calcium > 1000, positive MPS 20%. The age and gender adjusted calcium percentile score was also related to a positive MPS: < 50th percentile, positive MPS < 2%; > 50th percentile the incidence of a positive MPS increased, but did not exceed 11% even at the 90th percentile. Many of the 1119 normal MPS patients had positive CT scans: 78% of those with a > zero calcium score; 56% of those with > 100; 31% of those with > 400. Multivariate analysis showed that the log calcium score was the most potent predictor of a positive MPS vs other clinical variables (OR 3.4, CI 2.13-5.29, P < .0001). Berman et al concluded that: 1) an ischemic MPS is associated with a positive calcium scan, but rarely with a calcium score < 100; 2) a calcium score < 100 eliminates the need for a stress test; 3) normal MPS patients frequently have coronary calcium, suggesting a role for coronary calcium scanning in patients with a normal MPS if their risk for CAD is unclear.
Comment by Michael H. Crawford, MD
Despite the fact that few insurance companies reimburse for it, the use of fast CT to detect coronary calcium is increasing. Most of the increase has been from patient demand, as its acceptance by professional organizations has been lukewarm. This study by a group that has not been one of the advocates for fast CT, nor involved in its commercial exploitation, makes several important points for those of us who are confronted with patients bearing fast CT test results. First, the raw score has more predictive value than the age-sex adjusted calcium percentile score. Unfortunately, the patients are focused on the percentile score because it is usually high (> 50%), even if their raw score is relatively low, and percent numbers seem more understandable, even though they don’t mean that you have an X% chance of having a heart attack as soon as they believe. The percentile score does correlate with long-term risk and may help commit the patient to primary prevention measures for life. Second, there seems to be a threshold phenomenon with raw calcium scores. If they are < 100, almost no one has an ischemic MPS and it is not worth doing even if the percentile score is high. If it is > 400, there is a reasonable chance of getting a positive MPS even in asymptomatic patients. Scores between 100 and 400 are the grey zone where clinical judgment must be exercised. Asymptomatic patients in this range are less likely to have an ischemic MPS and could reasonably be managed conservatively. Third, stress testing does not reliably identify subclinical CAD. In this study, 88% of those with a negative MPS had detectable coronary calcium.
Even with calcium scores > 1000, the majority had a negative MPS. This emphasizes the point that has been made before that screening stress tests in asymptomatic patients are low yield and may give a false sense of security. Fast CT does detect subclinical disease and seems to be of incremental value over other traditional risk factors. Obviously, costs and logistics preclude the widespread use of CT screening, but in selected patients it can be of use. I find it particularly useful in asymptomatic patients with bad family histories to reassure or motivate them, depending on the results.
Dr. Crawford, Professor of Medicine, Associate Chief of Cardiology for Clinical Programs University of California San Francisco, is Editor of Clinical Cardiology Alert.