Stress/rest myocardial perfusion imaging (MPI) is commonly used for the diagnostic assessment of patients with suspected coronary artery disease (CAD) and for aiding clinical management decisions, such as the need for myocardial revascularization versus medical therapy. The latter use is facilitated by the well-documented relationship between the extent and severity of stress-induced myocardial perfusion abnormalities and the risk for subsequent adverse cardiac events. Despite these utilities, stress/rest MPI imaging has an important limitation: while myocardial perfusion abnormalities are useful for assessing the hemodynamic significance of obstructive coronary stenosis, many patients with normal stress/rest MPI studies have a substantial magnitude of subclinical coronary atherosclerosis, which cannot be detected because they are not flow-limiting in magnitude.^1,2 Coronary artery calcium (CAC) scanning is a complementary test that can detect subclinical atherosclerosis.

Over the years, CAC has been primarily used for screening purposes among asymptomatic individuals with risk factors for CAD. However, there is now growing recognition that CAC scanning could also serve as a clinical aid for improving the work-up and management of patients who are candidates for stress/rest MPI.^3,4

A number of features make CAC scanning make it an appealing test. The presence of CAC is a specific marker of CAD in an individual patient. Even very small amounts of CAC, with scores ranging 1–10, places patients at increased risk compared to patients with normal CAC scans.^5,6 Patho-anatomic studies conducted by Rumberger et al in the 1990s demonstrated a proportional relationship between the amount of CAC and total atherosclerotic burden.⁷ Since then, hundreds of outcome studies have consistently demonstrated a strong proportional relationship between the magnitude of CAC abnormality and the frequency of subsequent cardiac events. Most of these studies have been performed in asymptomatic populations. However, more recent studies have also demonstrated a similar proportional relationship between CAC scores and adverse outcomes among patients with chest pain symptoms. For example, in the recent Prospective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE) trial, a proportional relationship was also noted between the degree of CAC abnormality and the frequency of subsequent clinical events among patients with chest pain symptoms.⁸

For these reasons, there has been growing interest in evaluating the combined use of stress/rest MPI and CAC scanning for diagnostic and prognostic assessment. MPI using positron emission tomography (PET) is particularly well suited for this use because most PET scanners used for cardiac imaging today are hybrid PET/CT scanners, permitting both the anatomic assessment of coronary atherosclerosis and functional assessment of myocardial perfusion in a single study. In addition, hybrid SPECT/CT camera systems are now also increasingly available, although they presently constitute only a small percentage of the SPECT-MPI scanners. At the time of hybrid PET or SPECT imaging, the concomitant low dose non-ECG-gated CT scan is used to perform attenuation correction for the radionuclide images. This permits visual estimation of CAC abnormality, which in one formal study of 492 patients undergoing hybrid PET or SPECT/CT imaging was shown to have favorable agreement to the standard measurement of CAC abnormality during separately acquired ECG-gated calcium scans in the same patient population.⁹ Performance of SPECT-MPI with conventional cameras or stand-alone PET systems can also be combined with stand-alone CAC scanning to assess the same functional information that can be derived from hybrid scanners. Accordingly, this chapter reviews potential clinical applications that can be derived from the combined use of stress/rest MPI and CAC scanning. Second, we also examine how CAC scanning can be used as a triage method for optimizing the selection of patients for stress/rest MPI.

Studies performed in the early 2000s established a high frequency of atherosclerosis among patients with normal SPECT-MPI. For example, in a study involving 1195 patients that underwent both SPECT-MPI and CAC scanning, we noted that most patients with an abnormal SPECT-MPI study had CAC abnormality, but so too did 78% of 1119 patients with normal SPECT-MPI² (Figure 2-1). Among the nonischemic patients with CAC abnormality, nearly one-third had evidence of CAC scores >400, indicative of extensive subclinical atherosclerosis.

This observation of a high frequency of CAC abnormality among patients with normal SPECT-MPI studies is particularly notable in light of data indicating a declining frequency of abnormal SPECT-MPI studies over time. In our own experience, between 1991 and 2009 the frequency of ischemic SPECT-MPI studies fell from 29.6% to 5.0% among diagnostic patients who were referred for SPECT-MPI at Cedars-Sinai Medical Center¹⁰ (Figure 2-2). The presence of a temporal decline in inducible myocardial ischemia has also been noted in other large patient populations.^11,12 This has implications for patient management, because notwithstanding the important utilities of a normal MPI in guiding acute management decisions, it has generally had limited impact on guiding the intensity of CAD risk factor management. In contrast, concomitant use of CAC scanning with stress/rest MPI allows both improved prognostic risk assessment and opportunities for enhancing patient risk factor management based on the combined test results.

For years, the ability of radionuclide stress testing to predict myocardial risk was based on primarily short-term follow-up studies. These studies established that a normal SPECT study was associated with an excellent prognosis. For example, in a meta-analysis of 17 studies, Metz et al found that the annualized event rate associated with a normal exercise SPECT-MPI study was only 0.45% per year.¹³ The mean duration of follow-up in the studies constituting this meta-analysis was only 33.8 months. In recent years, however, there has been increasing interest in assessing the long-term outcomes associated with stress/rest MPI, in particular, among patients with normal MPI studies. These studies have demonstrated that long-term outcome among non-ischemic MPI patients is quite heterogeneous, influenced by a variety of clinical factors.

For example, in an extended follow-up of 5762 patients who had a normal stress-rest SPECT-MPI. Supariwala et al found that all-cause mortality increased according to both the number of CAD risk factors and whether stress testing was performed using pharmacologic or exercise stress.¹⁴ The frequency of deaths ranged from 0.8% per year for exercise patients who had no CAD risk factors to 4.2% per year among patients undergoing pharmacologic patients and having >two CAD risk factors, a 5-fold increase in risk (Figure 2-3). Similar findings were noted in another large study.¹⁵

In similar fashion, the concomitant use of CAC scanning can be used to improve long-term risk prediction among patients undergoing stress/rest MPI. CAC scanning provides complementary prognostic information both among patients with and without a myocardial perfusion abnormality. In this regard, Chang et al studied 1216 patients who were assessed by both SPECT-MPI and CAC scanning and followed for a median of 6.9 years.¹⁶ There was a progressive increase in adverse clinical events with increasing CAC score in patients with abnormal and those with normal SPECT-MPI. If both CAC scan and SPECT-MPI abnormalities were present, there was a synergistic elevation in patient risk (Figure 2-4). In similar fashion, Engbers et al assessed 4897 symptomatic patients without known CAD who underwent both stress/rest SPECT-MPI with CAC scoring.¹⁷ Adverse events included myocardial infarction, all-cause mortality or late revascularization. The median follow-up period was 940 days. These investigators also noted a stepwise increase in adverse clinical events with increasing CAC score if SPECT abnormality was present (Figure 2-5). Importantly, however, in patients with normal SPECT MPI, while the event rates were lower than in those with abnormal SPECT, there was an increased event rate according to the amount of CAC. A zero CAC scan has also been found to indicate a low likelihood of obstructive CAD when testing patients with a relatively low likelihood of CAD, but this association is not as reliable when applying CAC scanning to patients with a higher likelihood of CAD.^18,19

Among patients referred for stress/rest MPI for clinical reasons, the presence of myocardial ischemia is often a determining factor as to whether myocardial revascularization may be warranted. Further, once ischemia is present, aggressive medical management of CAD risk factors and clinical symptoms becomes warranted in all such patients. Among patients with a normal stress/rest MPI study, the intensity of medical management following testing may be considerably enhanced by considering the results of concomitant CAC scanning. Because any CAC abnormality among patients with a normal MPI puts them at increased long-term clinical risk,^5,6 this finding should signal the physician to intensity their clinical management of patients. The presence and extent of CAC among patients with normal MPI may influence the physician to more closely monitor and aggressively manage other CAD risk factors, such as initiating or increasing the intensity of statin therapy. In addition, knowledge of a high CAC score may be reason for the physician to encourage other lifestyle changes designed to reduce long-term clinical risk and to provide more feedback and/or frequent physician visits to assist patients. Further, for nonischemic patients with substantial CAC burden, the physician may choose to increase the frequency of follow-up or decrease the threshold for further testing. By contrast, a zero CAC score may be the reason for physicians to increase their threshold for further testing and reduce downstream medical testing.

Whereas, the actual impact of CAC scanning on influencing the behavior of physicians has not yet been studied among patients referred for stress/rest MPI, important insights in this regard can be garnered from the Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research (EISNER) trial, a prospective randomized trial that assessed the impact of CAC scanning on subsequent patient management and change in CAD risk factor profiles.²⁰ For this study, 2137 middle-aged volunteers with CAD risk factors were randomized in a 2:1 fashion into two groups: those receiving CAC scanning at baseline and those who were not scanned. A standard risk-factor counseling session was provided to both groups during an initial clinical visit supervised by a nurse practitioner, but the scan group was also counseled based on the results of CAC scanning. During 4-year follow-up, it became apparent that within the scan group that the intensity of medical risk-factor management, such as statin use and blood pressure medication, paralleled the CAC scores at the start of the study. The higher the CAC score, the greater was the use of these medications. Similarly, the frequency of downstream tests and procedures during follow-up paralleled the baseline magnitude of CAC abnormality. Among patients with a zero CAC score, the intensity of medical management and frequency of downstream medical resource utilization fell during follow-up compared to the frequency of medication or resource utilization that occurred in the group of subjects that were not referred for CAC scanning. This study thus shows that the results of CAC scanning can strongly influence physician management of patients.