Abstract
Background
Patients with a coronary artery calcium (CAC) score >300 on dedicated CAC scoring computed tomography (CT) are at equivalent risk of major adverse cardiac events (MACE) as those with established atherosclerotic cardiovascular disease (ASCVD).
Objectives
The aim of the study was to identify the extent of CAC on CT performed for attenuation correction (CTAC) as part of nuclear myocardial perfusion imaging that equates to secondary prevention.
Methods
We retrospectively studied 17,901 patients (48% female, age 64 ± 12 years, body mass index 30 kg/m2 [26-36]) who underwent nuclear myocardial perfusion imaging with CTAC (single photon emission computed tomography/CT or positron emission tomography/CT) at a single center. Prior ASCVD was defined as myocardial infarction (MI), cerebrovascular accident, peripheral artery disease, or prior revascularization. A semiquantitative visually estimated CAC score was obtained by scoring CAC in each coronary artery from 0 (absent) to 3 (severe), yielding a total score of 0 to 12 (zero, mild 1-2, moderate 3-6, and severe ≥7). The primary outcome was the composite of death, MI, or late revascularization.
Results
Among 13,852 patients without prior ASCVD (CAC zero 45%, mild 23%, moderate 21%, severe 11%) and 4,049 with ASCVD, 2,006 patients (11%) experienced MACE during a median follow-up of 25 (Q1-Q3: 10-43) months. In multivariable Cox regression, patients with severe calcification had no difference in risk for MACE, MI, or all-cause mortality vs ASCVD patients (P > 0.05).
Conclusions
Patients without prior ASCVD but with severe CAC (score ≥7) on CTAC demonstrated a risk for cardiovascular events and mortality comparable to those with known ASCVD, highlighting the need for more aggressive management in this high-risk primary prevention group.
Key words: ASCVD, attenuation correction CT, coronary artery calcification, major adverse cardiovascular events, primary prevention, secondary prevention
Central Illustration
Patients with atherosclerotic cardiovascular disease (ASCVD) have an increased risk for subsequent cardiovascular events and mortality.1 Significant lifestyle changes as well as pharmacotherapy with lipid-lowering drugs, and aspirin constitute the cornerstone of secondary prevention, with a progressively more aggressive approach based on the patient’s risk profile.2
Coronary artery calcification (CAC) is an important indicator of atherosclerosis in the coronary arteries and signifies the presence of coronary artery disease (CAD).3 The quantitative CAC score, expressed in Agatston units, can be calculated using cardiac electrocardiogram (ECG)-gated noncontrast computed tomography (CT) scans.4 A semiquantitative estimation of CAC burden can also be achieved by low-dose CT attenuation correction scans (CTAC) obtained during nuclear myocardial perfusion imaging (MPI).5, 6, 7 Traditional cardiovascular risk factors are strongly associated with the presence of calcification, and elevated CAC levels are linked to an increased risk of major adverse cardiac events (MACE) and mortality in both symptomatic and asymptomatic patients with CAD.8 Previously, it has been demonstrated that patients with CAC score above 300 on dedicated gated CT are at equivalent risk of MACE as those with established ASCVD.9
However, it remains unclear that among patients without established ASCVD, what visual CAC threshold on CTAC carries a comparable risk for subsequent MACE to those with established ASCVD.
Methods
Study design and data sources
This retrospective study used data from the institutional nuclear database derived from a single health system (Yale New Haven Health) with patients who underwent exercise or pharmacologic single photon emission computed tomography (SPECT) or positron emission tomography (PET) from July 2016 to May 2022. The Yale Institutional Research Ethics Board approved this study (Human Investigation Committee: 2000021621).
Study population and data collection
All patients who underwent a SPECT or PET MPI for CAD during the study period of July 1, 2016, to May 1, 2022, were screened (n = 40,370). We excluded patients who underwent SPECT or PET without CT (n = 20,184) and duplicate studies (2,285). For patients with repeated studies over the timeframe of our study, we only included their first MPI. The flow diagram of patient inclusion is presented in Supplemental Figure 1. The abstraction of data was performed by cardiology fellows and board-certified cardiologists on the day of the study. All relevant baseline demographic, clinical, medication information, and procedure-related information were collected. Follow-up information was obtained for all patients up to June 2022.
CT acquisition parameters
CTAC scans were performed free breathing without ECG-gating in helical mode. For PET, the following CTAC acquisition parameters were used with a 64 slice CT: tube current: 50 mA, tube voltage: 120 kV, rotation time: 0.4 seconds, pitch: 0.98, helical slice thickness: 3.75 mm, for patients with body mass index (BMI) ≥40 kg/m2, automated tube current modulation was used with tube current selected from a range of 50-150 mA. For CTAC acquired with SPECT, 2 set of acquisition parameters were used: 1) 64-slice CT for 72% of SPECT/CT cases: tube current: 60 mA, tube voltage: 120 kV, rotation time: 0.4 seconds, pitch: 0.98, helical slice thickness: 2.5 mm with 2.5 mm full angle reconstruction, for patients with BMI ≥40 kg/m2, automated tube current modulation was used with tube current selected from a range of 50-150 mA; and 2) 8-slice CT for 28% of SPECT/CT cases: tube current: 20 mA, tube voltage: 120 kV, rotation time: 0.4 seconds, pitch: 0.98, helical slice thickness: 5 mm with 5 mm standard reconstruction.
Data definitions and outcomes
The presence or absence and the extent of CAC was evaluated based on CTAC findings as previously described.6 The visual CAC score was determined by individually assessing calcification in each coronary artery (left main, left anterior descending, left circumflex artery, and right coronary artery) with a score of 0 (absent), 1 (mild), 2 (moderate), or 3 (severe) by experienced nuclear readers on the day of the study. These scores were then summed to produce a total score ranging from 0 to 12 for each patient, with categories defined as follows: 0 = no calcification, 1 to 2 = mild calcification, 3 to 6 = moderate calcification, and ≥7 = severe CAC. In a subset of patients, the numeric CAC score was also manually derived using the same CTAC acquisition. Outcomes were extracted from the electronic medical record. The clinical data sources for the present study population included the Yale Lumedx Imaging Reporting Database, the Yale Stress Registry, and the Yale New Haven Health System electronic medical records.
The primary outcome was MACE defined as all-cause mortality, nonfatal myocardial infarction (MI), or late coronary revascularization (percutaneous coronary intervention [PCI] or coronary artery bypass surgery [CABG] >90 days after SPECT or PET MPI). Secondary outcomes were nonfatal MI, all-cause mortality, and composite of all-cause mortality or nonfatal MI. Patients were considered to have a history of prior ASCVD if they had a documented nonfatal MI, cerebrovascular accident, peripheral artery disease, or prior coronary revascularization (PCI or CABG).
Statistical analysis
Categorical variables are presented as counts with proportions and the chi-squared test was used for comparisons. Distributions of continuous variables were assessed using histograms and distributional diagnostics (skewness, kurtosis). Continuous symmetric variables are presented as mean ± SD and compared using parametric tests (t-test), and skewed variables are presented as median (IQR) and compared using nonparametric tests (Mann-Whitney U). A univariate Cox proportional hazards model was used to identify the calcification threshold (from 0 to 12) of patients without prior ASCVD where the HR for MACE did not differ significantly (based on HRs near 1.0 with 95% CIs spanning 1.0) from patients with prior ASCVD. Kaplan-Meier survival curves were generated for the primary and secondary outcomes based on the presence or absence of prior ASCVD. Patients without prior ASCVD were further stratified by calcification severity (zero, mild, moderate, and severe). Univariate and multivariate Cox proportional hazards models were conducted to compare the risk of primary and secondary outcomes across different levels of calcification in patients without prior ASCVD, using patients with prior ASCVD as the reference group. Multivariable analysis was adjusted for sex, diabetes, hyperlipidemia, hypertension, family history of CAD, smoking status, renal disease, heart failure and myocardial perfusion, and stress left ventricular ejection fraction (LVEF) results. Covariates were selected based on clinical relevance and prior literature as potential confounders of the association between CAC category and outcomes. Collinearity among covariates was assessed using variance inflation factors, and no concerning collinearity was observed. Schoenfeld residual testing indicated nonproportionality in the adjusted models that were largely attributable to age and BMI. Adjusted models were refitted using a stratified Cox approach, stratifying baseline hazards by age and BMI quartiles, which resolved evidence of nonproportionality. To address multiple testing across CAC strata and outcomes, we performed a sensitivity analysis controlling the false discovery rate using the Benjamini-Hochberg procedure. To assess potential center-based variations, we conducted sensitivity survival analyses stratified by imaging centers, including centers with more than 1,000 patients available for evaluation. HRs are presented with 95% CIs. Correlation between the semiquantitative and numeric CAC scores was assessed using Spearman’s ρ and Kendall’s τ coefficients. Discrimination of the visual score for identifying quantitative CAC ≥300 was evaluated using receiver operating curve analysis. Statistical significance was assessed using a two-sided test with a P value threshold of 0.05. All analyses were conducted using STATA software (version 16; STATA Corporation).
Results
Patient characteristics
In total, 17,901 patients (48% female, mean age 64 ± 12 years, BMI: 30 kg/m2 [Q1-Q3: 26-36]) were included in the final analysis. Baseline patient characteristics are presented in Table 1. Of the 4,049 patients (23%) with a history of ASCVD, 1,394 (34%) had a prior MI, 805 (20%) had a prior cerebrovascular accident, 823 (20%) had a history of peripheral artery disease, and 2,267 (56%) had a history of prior revascularization (CABG or PCI). 13,852 patients (77%) had no prior ASCVD and were categorized in the following groups according to their calcification score: 6,335 (45%) with zero calcification, 3,157 (23%) with mild calcification (visual CAC score: 1-2), 2,873 (21%) with moderate calcification (visual CAC score: 3-6) and 1,487 (11%) with severe calcification (visual CAC score ≥7). Patients with prior ASCVD had a higher BMI, more comorbidities (heart failure, hypertension, and diabetes) and were more likely to have used medications including aspirin (76% vs 54%) and statins (82% vs 70%) compared with those without prior ASCVD and severe CAC (P < 0.001) (Table 1). In total, 5,421 patients (30%) underwent PET, of whom 1,683 (31%) had abnormal perfusion (Table 2). A total of 12,480 patients (70%) underwent SPECT, of whom 3,912 (31%) had abnormal perfusion (Table 2). The median stress LVEF was 65% (Q1-Q3: 56%-71%), and 1,432 (8%) of the total cohort had a positive exercise ECG test. Among patients without prior ASCVD, significant differences were observed across calcification groups in PET, SPECT, stress and rest myocardial blood flow, myocardial flow reserve, stress LVEF, and ECG test results (Table 2). In comparison to patients with ASCVD, patients with severe CAC had lower rate of abnormal SPECT perfusion, higher stress LVEF and higher rate of stress-induced ischemic ECG changes (P ≤ 0.002 for all). Details on perfusion results, stress and rest myocardial blood flow, myocardial flow reserve, and stress LVEF results are presented in Table 2.
Table 1.
Baseline Demographics and Clinical Characteristics of Patients
| Total (N = 17,901) | CAC Score Group |
Prior ASCVD (n = 4,049) | P Valuea | ||||
|---|---|---|---|---|---|---|---|
| None (0) (n = 6,335) | Mild (1-2) (n = 3,157) | Moderate (3-6) (n = 2,873) | Severe (≥7) (n = 1,487) | ||||
| Age, y, mean (SD) | 64 (12) | 57 (12) | 65 (10) | 69 (10) | 73 (10) | 68 (11) | <0.001 |
| BMI, kg/m2 | 30 (26-36) | 32 (27-39) | 30 (26-36) | 30 (26-35) | 29 (25-34) | 30 (26-34) | 0.002 |
| Female, n (%) | 8,666 (48%) | 3,963 (63%) | 1,650 (52%) | 1,197 (42%) | 472 (32%) | 1,384 (34%) | 0.08 |
| Renal disease, n (%) | 1,580 (9%) | 318 (5%) | 213 (7%) | 281 (10%) | 189 (13%) | 579 (14%) | 0.134 |
| Heart failure, n (%) | 1,476 (8%) | 295 (5%) | 185 (6%) | 253 (9%) | 163 (11%) | 580 (14%) | 0.001 |
| Hypertension, n (%) | 12,193 (68%) | 3,575 (56%) | 2,130 (67%) | 2,054 (71%) | 1,162 (78%) | 3,272 (81%) | 0.034 |
| Diabetes, n (%) | 4,879 (27%) | 1,411 (22%) | 764 (24%) | 810 (28%) | 480 (32%) | 1,414 (35%) | 0.071 |
| Hyperlipidemia, n (%) | 9,892 (55%) | 2,630 (42%) | 1,694 (54%) | 1,771 (62%) | 982 (66%) | 2,815 (70%) | 0.016 |
| Smoking, n (%) | 2,955 (17%) | 932 (15%) | 530 (17%) | 519 (18%) | 242 (16%) | 732 (18%) | 0.123 |
| Family history of CAD, n (%) | 2,368 (13%) | 1,035 (16%) | 479 (15%) | 369 (13%) | 142 (10%) | 343 (8%) | 0.203 |
| Prior myocardial infarction, n (%) | 1,394 (8%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 1,394 (34%) | NA |
| Prior cerebrovascular accident, n (%) | 805 (4%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 805 (20%) | NA |
| Prior peripheral vascular disease, n (%) | 823 (5%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 823 (20%) | NA |
| Prior revascularization (CABG or PCI), n (%) | 2,267 (13%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 2,267 (56%) | NA |
| Medications | |||||||
| Aspirin, n (%) | 8,345 (47%) | 1,963 (31%) | 1,190 (38%) | 1,321 (46%) | 804 (54%) | 3,067 (76%) | <0.001 |
| Clopidogrel, n (%) | 924 (5%) | 5 (0%) | 4 (0%) | 5 (0%) | 1 (0%) | 909 (22%) | <0.001 |
| Statin, n (%) | 10,181 (57%) | 2,396 (38%) | 1,640 (52%) | 1,780 (62%) | 1,048 (70%) | 3,317 (82%) | <0.001 |
| ACE inhibitors/ARBs, n (%) | 7,135 (40%) | 2,005 (32%) | 1,189 (38%) | 1,268 (44%) | 707 (48%) | 1,966 (49%) | 0.492 |
| Beta blockers, n (%) | 7,614 (43%) | 1,742 (27%) | 1,094 (35%) | 1,261 (44%) | 753 (51%) | 2,764 (68%) | <0.001 |
| Calcium channel blockers, n (%) | 4,746 (27%) | 1,323 (21%) | 794 (25%) | 877 (31%) | 505 (34%) | 1,247 (31%) | 0.026 |
| Nitrates, n (%) | 931 (5%) | 136 (2%) | 60 (2%) | 97 (3%) | 93 (6%) | 545 (13%) | <0.001 |
Bold values indicate statistical significance (P < 0.05).
ACE = angiotensin-converting enzyme; ARBs = angiotensin II receptor blockers; ASCVD = atherosclerotic cardiovascular disease; BMI = body mass index; CABG = coronary artery bypass grafting; CAC = coronary artery calcification; CAD = coronary artery disease; PCI = percutaneous coronary intervention.
The P value refers to comparison between severe (≥7) calcification and prior ASCVD groups.
Table 2.
Nuclear and ECG Test Results of Included Patients
| Total (N = 17,901) | Visual CAC Score Group |
P Valuea | Prior ASCVD (n = 4,049) | P Valueb | ||||
|---|---|---|---|---|---|---|---|---|
| None (0) (n = 6,335) | Mild (1-2) (n = 3,157) | Moderate (3-6) (n = 2,873) | Severe (≥7) (n = 1,487) | |||||
| Imaging test (PET/SPECT) | <0.001 | 0.012 | ||||||
| PET (total) | 5,421 (30%) | 1,914 (30%) | 835 (26%) | 845 (29%) | 452 (30%) | <0.001 | 1,375 (34%) | 0.669 |
| Normal perfusion, n (%) | 3,651 (67%) | 1,663 (87%) | 660 (79%) | 563 (67%) | 186 (41%) | 579 (42%) | ||
| Abnormal perfusion, n (%) | 1,683 (31%) | 208 (11%) | 162 (19%) | 269 (32%) | 263 (58%) | 781 (57%) | ||
| Equivocal or nondiagnostic, n (%) | 73 (1.3%) | 33 (1.7%) | 13 (1.5%) | 12 (1.4%) | 3 (0.6%) | 12 (0.9%) | ||
| Global MFR | 2.15 (IQR 1.70-2.64) | 2.42 (IQR 2.01-2.89) | 2.24 (IQR 1.82-2.73) | 2.05 (IQR 1.69-2.46) | 1.81 (IQR 1.45-2.18) | <0.001 | 1.86 (IQR 1.44-2.35) | 0.065 |
| Rest MBF (global) | 0.95 (IQR 0.78-1.22) | 0.99 (IQR 0.81-1.27) | 0.97 (IQR 0.79-1.27) | 0.95 (IQR 0.77-1.23) | 0.92 (IQR 0.76-1.14) | <0.001 | 0.90 (IQR 0.72-1.16) | 0.31 |
| Stress MBF (global) | 2.11 (IQR 1.61-2.70) | 2.45 (IQR 1.98-3.06) | 2.25 (IQR 1.84-2.79) | 1.97 (IQR 1.59-2.53) | 1.69 (IQR 1.31-2.11) | <0.001 | 1.71 (IQR 1.28-2.26) | 0.42 |
| SPECT (total) | 12,480 (70%) | 4,421 (70%) | 2,322 (74%) | 2,028 (71%) | 1,035 (70%) | <0.001 | 2,674 (66%) | <0.001 |
| Normal perfusion, n (%) | 8,358 (67%) | 3,707 (84%) | 1,743 (75%) | 1,301 (64%) | 510 (49%) | 1,097 (41%) | ||
| Abnormal perfusion, n (%) | 3,912 (31%) | 604 (14%) | 549 (24%) | 702 (35%) | 510 (49%) | 1,547 (58%) | ||
| Equivocal or nondiagnostic, n (%) | 179 (1.4%) | 87 (2%) | 28 (1.2%) | 24 (1.2%) | 13 (1.3%) | 27 (1%) | ||
| Stress LVEF | 65 (IQR 56-71) | 67 (IQR 59-73) | 66 (IQR 59-73) | 65 (IQR 56-71) | 62 (IQR 53-70) | <0.001 | 59 (IQR 48-68) | <0.001 |
| ECG response to test | <0.001 | 0.002 | ||||||
| Positive | 1,432 (8%) | 407 (7%) | 267 (9%) | 239 (8%) | 174 (12%) | 345 (9%) | ||
| Negative | 12,817 (74%) | 4,750 (77%) | 2,312 (74%) | 2,064 (73%) | 993 (67%) | 2,698 (71%) | ||
| Equivocal | 918 (5%) | 383 (6%) | 174 (6%) | 139 (5%) | 62 (4%) | 160 (4%) | ||
| Nondiagnostic | 2,442 (14%) | 633 (10%) | 377 (12%) | 404 (14%) | 249 (17%) | 779 (20%) | ||
Bold values indicate statistical significance (P < 0.05).
ASCVD = atherosclerotic cardiovascular disease; CAC = coronary artery calcification; ECG = electrocardiogram; LVEF = left ventricular ejection fraction; MBF = myocardial blood flow; MFR = myocardial flow reserve; PET = positron emission tomography; SPECT = single photon emission computed tomography.
Comparison between none (0), mild (1-2), moderate (3-6), and severe (≥7) calcification groups (ie, no-ASCVD group).
Comparison between severe (≥7) calcification and prior ASCVD groups.
Outcomes
In a median follow-up period of 25 (Q1-Q3: 10-43) months, 2,006 patients (11%) experienced MACE (691 MIs, 988 late revascularizations, and 1,194 deaths). On univariate Cox model for the outcome of MACE using prior ASCVD group as the reference, CAC score groups 0 to 6 were found to have significantly decreased risk for MACE (P < 0.05). Patients with CAC scores 7 to 11 were found to have similar risk for MACE (P > 0.05, HR close to 1.0 with 95% CI spanning 1.0), and patients with CAC score of 12 were found to have increased risk for MACE compared with prior ASCVD patients (HR: 1.75; 95% CI: 1.31-2.3) (Figure 1). Similar results were found for the outcomes of MI, all-cause mortality, and the composite of all-cause mortality and MI (Supplemental Table 1).
Figure 1.
Visual Coronary Artery Calcification Score and Major Adverse Cardiovascular Events
HRs for major adverse cardiac events by visual CAC score (0-12) in patients without prior ASCVD, with prior ASCVD as the reference. CAC scores 7 to 11 showed risk comparable to prior ASCVD, while score 12 was associated with higher risk. ASCVD = atherosclerotic cardiovascular disease; CAC = coronary artery calcification.
Survival analysis and univariate Cox model revealed that patients without prior ASCVD with zero, mild (CAC score: 1-2), or moderate (CAC score: 3-6) calcification had significantly lower risk for MACE, MI, all-cause mortality, and composite of all-cause mortality and MI compared with prior ASCVD patients (P < 0.05) (Table 3, Figure 2). Patients without prior ASCVD but with severe calcification (CAC score ≥7) showed no significant difference in risk for MACE (HR: 0.99; 95% CI: 0.85-1.12; P = 0.842), all-cause mortality (HR: 1.16; 95% CI: 0.96-1.37; P = 0.088), or the composite endpoint of all-cause mortality and MI (HR: 0.99; 95% CI: 0.84-1.14; P = 0.904), and a significantly lower risk of MI (HR: 0.76; 95% CI: 0.58-0.95; P = 0.028), compared with ASCVD patients (Table 3).
Table 3.
Univariate and Multivariate Cox Models for the Primary and Secondary Outcomes of Different Calcification Groups (None, Mild, Moderate, and Severe) Using Patients With Prior ASCVD as the Reference Group
| Patient Group | Univariate |
Multivariate |
||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P Value | aHR | 95% CI | P Value | |
| MACE | ||||||
| Prior ASCVD | Ref. | NA | NA | Ref. | NA | NA |
| CAC = 0 (none) | 0.19 | 0.16–0.21 | <0.001 | 0.32 | 0.27–0.37 | <0.001 |
| CAC = 1–2 (mild) | 0.33 | 0.28–0.38 | <0.001 | 0.49 | 0.42–0.57 | <0.001 |
| CAC = 3–6 (moderate) | 0.56 | 0.49–0.63 | <0.001 | 0.71 | 0.62–0.81 | <0.001 |
| CAC ≥7 (severe) | 0.99 | 0.85–1.12 | 0.842 | 0.99 | 0.86–1.14 | 0.889 |
| MI | ||||||
| Prior ASCVD | Ref. | NA | NA | Ref. | NA | NA |
| CAC = 0 (none) | 0.17 | 0.13–0.21 | <0.001 | 0.35 | 0.26–0.45 | <0.001 |
| CAC = 1–2 (mild) | 0.28 | 0.21–0.35 | <0.001 | 0.51 | 0.39–0.66 | <0.001 |
| CAC = 3–6 (moderate) | 0.46 | 0.36–0.56 | <0.001 | 0.66 | 0.53–0.83 | <0.001 |
| CAC ≥7 (severe) | 0.76 | 0.58–0.95 | 0.028 | 0.82 | 0.64–1.05 | 0.112 |
| All-cause mortality | ||||||
| Prior ASCVD | Ref. | NA | NA | Ref. | NA | NA |
| CAC = 0 (none) | 0.24 | 0.20–0.29 | <0.001 | 0.44 | 0.36–0.53 | <0.001 |
| CAC = 1–2 (mild) | 0.43 | 0.35–0.51 | <0.001 | 0.63 | 0.52–0.77 | <0.001 |
| CAC = 3–6 (moderate) | 0.65 | 0.54–0.75 | <0.001 | 0.79 | 0.66–0.94 | 0.007 |
| CAC ≥7 (severe) | 1.16 | 0.96–1.37 | 0.088 | 1.1 | 0.91–1.31 | 0.321 |
| All-cause mortality and MI | ||||||
| Prior ASCVD | Ref. | NA | NA | Ref. | NA | NA |
| CAC = 0 (none) | 0.22 | 0.19–0.26 | <0.001 | 0.4 | 0.34–0.48 | <0.001 |
| CAC = 1–2 (mild) | 0.38 | 0.32–0.44 | <0.001 | 0.59 | 0.50–0.69 | <0.001 |
| CAC = 3–6 (moderate) | 0.59 | 0.51–0.67 | <0.001 | 0.75 | 0.65–0.87 | <0.001 |
| CAC ≥7 (severe) | 0.99 | 0.84–1.14 | 0.904 | 0.98 | 0.84–1.14 | 0.783 |
Bold values indicate statistical significance (P < 0.05).
aHR = adjusted hazard ratio; ASCVD = atherosclerotic cardiovascular disease; CAC = coronary artery calcification; MACE = major adverse cardiovascular events; MI = myocardial infarction; NA = not applicable.
Figure 2.
Major Adverse Cardiovascular Events by Visual Coronary Artery Calcification Category
Kaplan-Meier curves of patients without prior ASCVD stratified by calcification level and patients with prior ASCVD (calcification groups represent patients without prior ASCVD). Abbreviations as in Figure 1.
Similar findings were observed using a multivariate model after adjusting for clinical and radiologic confounders. Patients with severe calcification (CAC score ≥7) showed no significant difference in risk for MACE (adjusted HR [aHR]: 0.99; 95% CI: 0.86-1.14; P = 0.889), MI (aHR: 0.82; 95% CI: 0.64-1.05; P = 0.112), all-cause mortality (aHR: 1.1; 95% CI: 0.91-1.31; P = 0.321), or the composite endpoint of all-cause mortality and MI (aHR: 0.98; 95% CI: 0.84-1.14; P = 0.783), compared with ASCVD patients (Table 3). After Benjamini-Hochberg false discovery rate correction, the pattern of findings and overall conclusions were unchanged (Supplemental Tables 2 and 3).
Sensitivity analyses stratified by imaging center (2 SPECT centers and 1 PET center) showed findings consistent with the primary analysis. Across centers, CAC 0, mild (CAC 1-2), and moderate CAC (CAC 3-6; except for 1 center) were consistently associated with substantially lower MACE risk vs prior ASCVD, while CAC ≥7 demonstrated adjusted HRs with 95% CIs spanning 1.0, supporting agreement of the severe-CAC/prior ASCVD risk comparability across different sites (Supplemental Figure 2).
When severe CAC group was subdivided into 7 to 11 and 12, the 7 to 11 subgroup remained comparable to prior ASCVD for MACE, death, and death/MI. In contrast, CAC = 12 was associated with higher risk for MACE (aHR: 1.49; 95% CI: 1.11-2), death (aHR: 1.62; 95% CI: 1.13-2.32), and the composite of death or MI (aHR: 1.59; 95% CI: 1.17-2.16) (Supplemental Table 4).
Additionally, evaluating late coronary revascularization as a separate outcome demonstrated a consistent risk pattern, with severe CAC showing comparable risk to those of prior ASCVD (unadjusted HR: 0.99; 95% CI: 0.86-1.13; aHR: 0.98, 95% CI: 0.86-1.13).
Abnormal perfusion on MPI testing was associated with elevated risk for the primary and secondary outcomes both in the univariate and the multivariate Cox regression model (Supplemental Table 5).
In a paired subset with both semiquantitative visual CAC and quantitative CAC derived from the same CTAC acquisition (n = 4,378), visual CAC correlated strongly with quantitative CAC (Spearman ρ = 0.885; Kendall’s τ-b = 0.756; both P < 0.001) (Supplemental Figure 3).
The visual score discriminated quantitative CAC ≥300 with an area under the receiver operating characteristic curve of 0.94 (95% CI: 0.93-0.95) (Supplemental Figure 4).
Aspirin and statin use
Data on aspirin and statin use at the time of stress MPI were available for all patients (17,901) (Table 1). Prescription rates within 90 days after the imaging study were only available for 11,300 patients (63% of the total cohort) and are presented in Supplemental Table 6. At 90 days after stress MPI, patients with prior ASCVD had significantly higher prescription rates than those with severe CAC for both statins (87% vs 78%) and aspirin (79% vs 60%) (P < 0.001).
Discussion
In this large cohort study of patients undergoing SPECT/CT or PET/CT MPI, the presence of increasing CAC burden evaluated on attenuation-correction CT was associated with an increased risk of cardiovascular events and mortality. Among patients without a known history of ASCVD, those with a semiquantitative visual CAC score greater than or equal to 7, derived from CTAC, were found to have a similar adjusted risk for MACE, MI, all-cause mortality, and the composite of all-cause mortality and MI, compared with patients with established atherosclerotic disease (Central Illustration).
Central Illustration.
Severe Computed Tomography Attenuation Correction Scans Coronary Artery Calcification Signals Atherosclerotic Cardiovascular Disease-Level Risk
Among 17,901 patients undergoing SPECT/CT or PET/CT myocardial perfusion imaging with CT attenuation correction, semiquantitative coronary artery calcification (CAC) was scored visually in each artery (left main, left anterior descending, left circumflex, and right coronary artery) as 0 = none, 1 = mild, 2 = moderate, and 3 = severe, then summed to a total score of 0 to 12. CAC categories were 0, 1 to 2, 3 to 6, and ≥7. Patients without prior ASCVD but severe CAC (score ≥7) had major adverse cardiac risk comparable to those with established ASCVD, supporting consideration of secondary prevention-level management. CT = computed tomography; PET = positron emission tomography; SPECT = single photon emission computed tomography; other abbreviations as in Figure 1.
This similarity in risk was supported by the overlap of Kaplan-Meier curves and in univariate and multivariate Cox models, where the HRs for patients with severe calcification (CAC score ≥7) were not significantly different from those of the prior ASCVD group, with HR estimates being close to 1 and 95% CIs spanning 1.0, suggesting clinically comparable risk. Of note, within the severe CAC category, patients with a visual score of 12 had higher risks of MACE, mortality, and the mortality/MI composite, than those with prior ASCVD, suggesting that this subgroup may represent an “extreme” calcification phenotype beyond severe CAC, consistent with the CAD-RADS 2.0 description of extensive plaque burden (CAC >1,000 Agatston units).10 According to our definition, a visual CAC score of 7 can be achieved through various combinations of coronary calcification, including: 2 severely and 1 mildly calcified coronary artery; 1 severely and 2 moderately calcified arteries; or 1 severely, 1 moderately, and 2 mildly calcified arteries.
The results of our study indicate that patients with severe coronary calcification should potentially be managed as aggressively as patients receiving secondary prevention. The 2018 American College of Cardiology/American Heart Association Guideline on the Management of Blood Cholesterol recommends high-intensity statin therapy to lower low-density lipoprotein cholesterol by ≥50% in patients with clinical ASCVD and the use of nonstatin therapies, such as ezetimibe or PCSK9 inhibitors in high-risk patients who do not meet low-density lipoprotein cholesterol targets.2 Guidelines also recommend that CAC assessment could be helpful when the decision to initiate statin therapy remains uncertain in patients without ASCVD (primary prevention). Specifically, a CAC score of ≥100 Agatston units or a score at or above the 75th percentile is recommended as a threshold to guide the initiation of statin therapy.2 However, unlike secondary prevention, the guidelines do not provide specific recommendations for intensified lipid-lowering treatment such as high-intensity statins or additional nonstatin therapies for primary prevention patients with severe calcification scores. Likewise, while the use of aspirin and other antiplatelet agents is well established and recommended for secondary prevention,11, 12, 13 its role in primary prevention remains uncertain, with clinical trials demonstrating moderate or minimal benefit, while also demonstrating increased risk of bleeding.12,14, 15, 16, 17, 18 In our cohort, patients with severe calcification were less likely to be on statins and aspirin (or other antiplatelets) compared with secondary prevention patients at baseline (statin use: 70% vs 82%; aspirin use: 54% vs 76%; P < 0.001 for both). Similar patterns were seen in the prescription rates within 90 days after the imaging study (statin use: 78% vs 87%; aspirin use: 60% vs 79%; P < 0.001 for both), suggesting a subgroup in primary prevention who could have potentially benefited from the initiation or intensification of lipid-lowering and/or antiplatelet therapy. Indeed, a previous study by Ajufo et al demonstrated that among primary prevention patients with low bleeding risk and a borderline (>5%) 10-year ASCVD risk, a CAC score threshold of 100 identified individuals who would experience net benefit (fewer ASCVD events with acceptable bleeding risk) versus those unlikely to benefit.19 Beyond lipid-lowering therapies, patients with severe CAC scores may benefit from aggressive risk factor modification, including lifestyle changes, blood pressure control, and smoking cessation, similar to secondary prevention strategies. Importantly, although patients with mild and moderate CAC in our cohort had lower event rates than the prior ASCVD reference group, this should not be interpreted as low cardiovascular risk. Large population studies have consistently demonstrated a graded relationship between CAC burden and incident coronary heart disease, with event risk increasing even at low CAC levels compared with zero CAC.8 Current prevention guidelines also support intensifying preventive therapy in the presence of CAC, particularly when CAC is ≥100 Agatston units in standard CAC scoring systems.2
Our results are in accordance with a previous study by Budoff et al, in which traditional ECG-gated noncontrast CT scans were used.9 The authors reported similar risk for MACE and mortality between patients with a CAC score above 300 Agatston units and those with prior ASCVD. Notably, in our study, the incremental risk associated with calcification surpassed that of prior ASCVD patients when calcification score reached 12 (ie, severe calcification across all 4 measured coronary artery territories). It is unknown whether this was also observed in the study by Budoff et al, since no data were reported for patients with more extensive calcification. In a study by Dzaye et al, data from 20,207 primary prevention patients with a 10-year ASCVD risk of ≥7.5% (sourced from the CAC Consortium) were analyzed.20 The study compared annualized ASCVD mortality rates for these patients with those in the secondary prevention population from the FOURIER trial (Further Cardiovascular Outcomes Research with PCSK9 Inhibition in Subjects with Elevated Risk).21 Findings revealed that CAC scores above 775 in the general at-risk primary prevention population, and scores above 300 for individuals with diabetes, corresponded to ASCVD mortality risks similar to those observed in the secondary prevention population from the FOURIER trial.20
In our study a semiquantitative CAC score was used, derived from low-dose attenuation correction CT obtained during SPECT/CT and PET/CT. This hybrid approach of using CTAC with MPI provides valuable prognostic information beyond that of perfusion imaging alone.6 Previous evidence supports a good correlation and agreement and comparable prognostic value for MACE between this method and the dedicated CAC scoring CT.5,6,22 With high interobserver reproducibility, semiquantitative visual estimation of the CAC score offers additional benefits, including reduced scan time, cost, and radiation exposure.6 Beyond visual scoring, several artificial intelligence systems now quantify CAC on ECG-gated and routine nongated chest CT (including CTAC for SPECT/PET) with good agreement to expert and gated-CT references. However, their performance may vary by vendor and reconstruction kernel, often requiring site-specific calibration and local validation before clinical use.23, 24, 25
Study Limitations
Despite the relatively large sample size, our study has significant limitations related to the single-center, retrospective study design. First the study population consisted of patients evaluated for suspected CAD, which may introduce selection bias and limit the generalizability of the findings to other groups given that those patients were probably more likely to have significant CAC compared with the general population, given their comorbidities. Second, the assessment of semiquantitative calcium score is more prone to inter- and intra-observer variabilities compared with traditional calcium score. In this study, calcium scores were calculated from the same non-ECG-gated attenuation CT examinations, therefore calcium scores may have been underestimated or overestimated. Third, data on trends in prescribed pharmacotherapy following the imaging study were only partially available, limiting postimaging comparison between patients with severe calcification and those with prior ASCVD. Finally, the follow-up time was relatively short and may underestimate late events. Overall, it remains uncertain whether these outcomes can be generalized, or the approach adopted in clinical practice to produce meaningful medication changes in many patients.
Conclusions
Patients without prior ASCVD who had severe CAC as suggested by a semiquantitative CAC score of ≥7 derived from low-dose CTAC scans during nuclear MPI, exhibited a MACE risk similar to that of patients with established ASCVD. Future research is needed to explore how CTAC for CAC assessment as part of a comprehensive nuclear stress test can improve the study quality and lead to more robust decisions regarding downstream testing but also potentially guide future lifestyle and pharmacotherapy strategies for patients without established ASCVD.
Funding support and author disclosures
This research was supported in part by grant R35HL161195 from the National Heart, Lung, and Blood Institute of the National Institutes of Health (Principal Investigator: Dr Slomka). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Footnotes
The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the Author Center.
Appendix
For supplemental tables and figures, please see the online version of this paper.
Supplemental material
References
- 1.Writing Committee M., Virani S.S., Newby L.K., et al. 2023 AHA/ACC/ACCP/ASPC/NLA/PCNA guideline for the management of patients with chronic coronary disease: a report of the American Heart Association/American College of Cardiology Joint Committee on clinical practice guidelines. J Am Coll Cardiol. 2023;82(9):833–955. doi: 10.1016/j.jacc.2023.04.003. [DOI] [PubMed] [Google Scholar]
- 2.Grundy S.M., Stone N.J., Bailey A.L., et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of blood cholesterol: executive summary: a report of the American College of Cardiology/American Heart Association task force on clinical practice Guidelines. J Am Coll Cardiol. 2019;73(24):3168–3209. doi: 10.1016/j.jacc.2018.11.002. [DOI] [PubMed] [Google Scholar]
- 3.Mori H., Torii S., Kutyna M., Sakamoto A., Finn A.V., Virmani R. Coronary artery calcification and its progression: what does it really mean? JACC Cardiovasc Imaging. 2018;11(1):127–142. doi: 10.1016/j.jcmg.2017.10.012. [DOI] [PubMed] [Google Scholar]
- 4.Agatston A.S., Janowitz W.R., Hildner F.J., Zusmer N.R., Viamonte M., Jr., Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol. 1990;15(4):827–832. doi: 10.1016/0735-1097(90)90282-t. [DOI] [PubMed] [Google Scholar]
- 5.Einstein A.J., Johnson L.L., Bokhari S., et al. Agreement of visual estimation of coronary artery calcium from low-dose CT attenuation correction scans in hybrid PET/CT and SPECT/CT with standard Agatston score. J Am Coll Cardiol. 2010;56(23):1914–1921. doi: 10.1016/j.jacc.2010.05.057. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Feher A., Pieszko K., Shanbhag A., et al. Comparison of the prognostic value between quantification and visual estimation of coronary calcification from attenuation CT in patients undergoing SPECT myocardial perfusion imaging. Int J Cardiovasc Imaging. 2024;40(1):185–193. doi: 10.1007/s10554-023-02980-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Mylonas I., Kazmi M., Fuller L., et al. Measuring coronary artery calcification using positron emission tomography-computed tomography attenuation correction images. Eur Heart J Cardiovasc Imaging. 2012;13(9):786–792. doi: 10.1093/ehjci/jes079. [DOI] [PubMed] [Google Scholar]
- 8.Detrano R., Guerci A.D., Carr J.J., et al. Coronary calcium as a predictor of coronary events in four racial or ethnic groups. N Engl J Med. 2008;358(13):1336–1345. doi: 10.1056/NEJMoa072100. [DOI] [PubMed] [Google Scholar]
- 9.Budoff M.J., Kinninger A., Gransar H., et al. When does a calcium score equate to secondary prevention?: insights from the multinational CONFIRM registry. JACC Cardiovasc Imaging. 2023;16(9):1181–1189. doi: 10.1016/j.jcmg.2023.03.008. [DOI] [PubMed] [Google Scholar]
- 10.Cury R.C., Leipsic J., Abbara S., et al. CAD-RADS 2.0 - 2022 coronary artery disease-reporting and data system: an expert consensus document of the Society of Cardiovascular Computed Tomography (SCCT), the American College of Cardiology (ACC), the American College of Radiology (ACR), and the North America Society of Cardiovascular Imaging (NASCI) J Cardiovasc Comput Tomogr. 2022;16(6):536–557. doi: 10.1016/j.jcct.2022.07.002. [DOI] [PubMed] [Google Scholar]
- 11.Vandvik P.O., Lincoff A.M., Gore J.M., et al. Primary and secondary prevention of cardiovascular disease: Antithrombotic therapy and prevention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2012;141(2 Suppl):e637S–e687S. doi: 10.1378/chest.11-2306. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Visseren F.L.J., Mach F., Smulders Y.M., et al. 2021 ESC guidelines on cardiovascular disease prevention in clinical practice. Eur Heart J. 2021;42(34):3227–3337. doi: 10.1093/eurheartj/ehab484. [DOI] [PubMed] [Google Scholar]
- 13.Tasoudis P.T., Kyriakoulis I.G., Sagris D., Diener H.C., Ntaios G. Clopidogrel monotherapy versus aspirin monotherapy in patients with established cardiovascular disease: systematic review and meta-analysis. Thromb Haemost. 2022;122(11):1879–1887. doi: 10.1055/a-1853-2952. [DOI] [PubMed] [Google Scholar]
- 14.Abdelaziz H.K., Saad M., Pothineni N.V.K., et al. Aspirin for primary prevention of cardiovascular events. J Am Coll Cardiol. 2019;73(23):2915–2929. doi: 10.1016/j.jacc.2019.03.501. [DOI] [PubMed] [Google Scholar]
- 15.Arnett D.K., Blumenthal R.S., Albert M.A., et al. 2019 ACC/AHA guideline on the primary prevention of cardiovascular disease: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines. Circulation. 2019;140(11):e596–e646. doi: 10.1161/CIR.0000000000000678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gaziano J.M., Brotons C., Coppolecchia R., et al. Use of aspirin to reduce risk of initial vascular events in patients at moderate risk of cardiovascular disease (ARRIVE): a randomised, double-blind, placebo-controlled trial. Lancet. 2018;392(10152):1036–1046. doi: 10.1016/S0140-6736(18)31924-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Group A.S.C., Bowman L., Mafham M., et al. Effects of aspirin for primary prevention in persons with diabetes mellitus. N Engl J Med. 2018;379(16):1529–1539. doi: 10.1056/NEJMoa1804988. [DOI] [PubMed] [Google Scholar]
- 18.McNeil J.J., Wolfe R., Woods R.L., et al. Effect of aspirin on cardiovascular events and bleeding in the healthy elderly. N Engl J Med. 2018;379(16):1509–1518. doi: 10.1056/NEJMoa1805819. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ajufo E., Ayers C.R., Vigen R., et al. Value of coronary artery calcium scanning in association with the net benefit of aspirin in primary prevention of atherosclerotic cardiovascular disease. JAMA Cardiol. 2021;6(2):179–187. doi: 10.1001/jamacardio.2020.4939. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Dzaye O., Razavi A.C., Michos E.D., et al. Coronary artery calcium scores indicating secondary prevention level risk: findings from the CAC consortium and FOURIER trial. Atherosclerosis. 2022;347:70–76. doi: 10.1016/j.atherosclerosis.2022.02.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sabatine M.S., Giugliano R.P., Keech A.C., et al. Evolocumab and clinical outcomes in patients with cardiovascular disease. N Engl J Med. 2017;376(18):1713–1722. doi: 10.1056/NEJMoa1615664. [DOI] [PubMed] [Google Scholar]
- 22.Isgum I., de Vos B.D., Wolterink J.M., et al. Automatic determination of cardiovascular risk by CT attenuation correction maps in Rb-82 PET/CT. J Nucl Cardiol. 2018;25(6):2133–2142. doi: 10.1007/s12350-017-0866-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Eng D., Chute C., Khandwala N., et al. Automated coronary calcium scoring using deep learning with multicenter external validation. NPJ Digit Med. 2021;4(1):88. doi: 10.1038/s41746-021-00460-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Miller R.J.H., Pieszko K., Shanbhag A., et al. Deep learning coronary artery calcium scores from SPECT/CT attenuation maps improve prediction of major adverse cardiac events. J Nucl Med. 2023;64(4):652–658. doi: 10.2967/jnumed.122.264423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.van Velzen S.G.M., Lessmann N., Velthuis B.K., et al. Deep learning for automatic calcium scoring in CT: validation using multiple cardiac CT and chest CT protocols. Radiology. 2020;295(1):66–79. doi: 10.1148/radiol.2020191621. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.