2023 Volume 30 Issue 10 Pages 1289-1302
Coronary artery calcium (CAC), which is detected using computed tomography scanning, is a well-established indicator of subclinical atherosclerosis. The CAC score is independently associated with atherosclerotic cardiovascular disease (ASCVD) outcomes and provides improved predictive values for estimating the risk of ASCVD beyond traditional risk factors. Thus, CAC is considered to have important implications for reclassification as a decision aid among individuals in the preclinical phase and as the primary prevention of ASCVD. This review is focused on epidemiological evidence on CAC in asymptomatic population-based samples from Western countries and Japan. We also discuss the usability of CAC as a tool for assessing ASCVD risk and its role in the primary prevention of ASCVD. A lack of evidence for the CAC score in ASCVD risk assessment beyond traditional risk factors in populations other than those in Western countries (including Japan) warrants further investigation. Clinical trials are also necessary to demonstrate the usefulness and safety of CAC screening in the primary prevention of ASCVD.
Atherosclerotic cardiovascular disease (ASCVD), including coronary heart disease (CHD), remains a major cause of death worldwide and accounts for over 30% of annual global fatalities1). ASCVD is also the leading cause of disease burden globally. Prevalent cases of total ASCVD nearly doubled from 271 million in 1990 to 523 million in 2019, and the number of ASCVD deaths has continuously increased from 12.1 million in 1990 to 18.6 million in 2019 2). Nonetheless, the burden of CVD remains unchanged, owing to an increasing number of patients surviving into old age with debilitating cardiovascular problems3). Additionally, more than half of deaths due to CHD occur outside the hospital, and approximately 50% of these are sudden4). In Japan, the incidence of CHD, including acute myocardial infarction, has recently shown an increasing trend5, 6). Therefore, primary prevention and risk stratification of ASCVD, including CHD, among asymptomatic adults are critical in clinical practice and public health.
Screening asymptomatic individuals using atherosclerosis in its subclinical phase to prevent ASCVD is currently the subject of intensive research7). Atherosclerosis begins in early life and accumulates silently until clinical symptoms appear late in the course of the clinical disease. Histopathological studies have shown that the extent of atherosclerosis is associated with traditional risk factors for ASCVD8, 9). Therefore, evaluating subclinical atherosclerosis may have important implications for understanding the accumulated burden of cardiovascular risk factors during the long-term latent period prior to clinical disease. Additionally, assessing subclinical atherosclerosis may be important for intervention in the preclinical phase, such as for reclassification, to prevent ASCVD that occurs in later life.
Currently, various noninvasive measures are available for the assessment of subclinical atherosclerosis, including coronary artery calcium (CAC), aortic artery calcium, carotid intima–media thickness and plaque, ankle–brachial index, and aortic pulse wave velocity. In this review, we focus on CAC, which is widely available, exhaustively studied, and a highly validated measure of subclinical atherosclerotic burden in coronary arteries. We also review the epidemiologic data on CAC in asymptomatic population-based samples from Western countries and Japan. Finally, we discuss the usability of CAC as a tool for assessing ASCVD risk and its role in the primary prevention of ASCVD.
Measurement of CACFor decades, quantification of CAC using computed tomography (CT) has continued to improve and has become a standardized measure of subclinical atherosclerosis. In 1990, Agatston et al. developed the first practically applicable quantitative CAC protocol10). This score is obtained as a weighted sum of the area of calcification (≥ 1 mm2) with its density, measured in Hounsfield units, with different weights for different densities (denser calcification has greater Hounsfield units and thus has a greater weight; Fig.1)10). Agatston’s CAC score requires a relatively complex measurement technique. In an effort to simplify the measurement of CAC and increase its reproducibility, the volume score was introduced by Callister et al.11). This score is simply calculated using the segmented calcified plaque area and the number of slices containing each of those plaques. Although conceptually different, the two CAC scoring methods share many similarities and produce highly correlated results12). With respect to their ability to rank order scores, the interclass correlation coefficient between the two methods was ＞0.95. Although some differences have been described, for example, a greater interscan reproducibility of the volume score compared with the Agatston score, these differences are not thought to be of sufficient magnitude to change clinical practice12). In this context, because it came first, the Agatston score remains the gold standard for CAC scoring. Therefore, in this review, CAC scoring is exclusively defined as CAC scoring using the Agatston method. Histopathological studies in humans have shown that CAC identified using CT is strongly and positively correlated with coronary plaques measured on autopsy. However, the calcium area is one-fifth or smaller than the plaque area13) and is poorly related to the severity of luminal stenosis14). Additionally, noncalcified or partially calcified plaques with a large necrotic core and expansive remodeling more frequently lead to myocardial infarction15). Owing to the strong association between the calcium area and plaque area in the coronary artery, the amount of CAC is considered to represent the overall magnitude of the atherosclerotic burden14).
On a CT image, calcification is expressed as a white lesion (≥ 130 HU). Ao, aorta; LA, left atrium.
The exact mechanism of calcium deposition in atherosclerotic plaque remains to be elucidated. However, many researchers consider it an active process involving arterial osteoblasts and osteoclasts16). This is in contrast to a passive process of calcification occurring in arterial media layers known as Mönckeberg’s sclerosis, which is associated with advanced tissue degeneration17). Intriguingly, statins, which are the mainstay of therapy for individuals at risk of ASCVD and definitively reduce the risk of clinical ASCVD events, are not associated with a reduction in calcium scores. This has been assessed in several randomized controlled clinical trials, each demonstrating that despite a significant low-density lipoprotein (LDL)-lowering effect, either CAC scores do not change or they trend toward higher levels in patients treated with statins18-21). The explanation for this apparent paradox with statin treatment remains an area of debate and investigation. The exact mechanisms by which statins may affect vascular calcification are not fully understood, but statin treatment may change the localization of calcium microdeposits, promoting their accumulation around the necrotic core to stabilize plaques22).
Comparison of CAC burden between the United States and JapanCompared with the United States (US) and other developed countries, Japan has had a much lower rate of CHD mortality, and this has largely been attributed to population-wide lower concentrations of serum total cholesterol23, 24). The current comparison data still support this specific feature of CHD mortality in Japan25). Recently, however, levels of several coronary risk factors have become comparable between the US and Japan, particularly among men24). Serum total cholesterol levels in Japanese men have steadily increased and reached levels similar to those observed in US men24). Additionally, Japanese men have a similar or higher prevalence of hypertension, diabetes mellitus, and cigarette smoking compared with US men24). Thus, some epidemiologic studies in Japan have observed a trend of increasing CHD incidence among men due to less favorable coronary risk factor profiles5, 6), whereas overall CHD incidence in the US appears to have declined in recent years26-28). Given these changing trends between the US and Japan, it is of interest to investigate whether a difference remains in the burden of coronary atherosclerosis between the two countries and, if so, whether this difference could be explained by a discrepancy in the distributions of traditional coronary risk factors. Therefore, comparing CAC scores is likely to offer investigators the opportunity to gain insight into the overall burden of subclinical atherosclerosis.
International comparisons using well-standardized methods can be made to investigate differences in the distribution of diseases and the risk factors responsible for the differences. This can increase understanding of the genetic and environmental origins of disease, which could result in disease prevention29). The Electron-Beam Tomography, Risk Factor Assessment Among Japanese and US Men in the Post-World War II Birth Cohort (ERA-JUMP) study investigated asymptomatic Japanese men in Japan, White men, and Japanese American men in the US in their 40s30-32). The study showed that Japanese men in Japan had significantly lower levels of coronary atherosclerosis, detected using CAC, than White or Japanese American men in the US. However, Japanese American men in the US had similar or higher levels of CAC compared with White men in the US. The prevalence of CAC scores ≥ 10 among Japanese men in Japan, White men, and Japanese American men in the US was 9.3%, 26.1%, and 31.4%, respectively30-32). These findings indicate that differing lifestyle factors, such as diet, but not genetic factors, contribute to the difference in coronary atherosclerosis between Japan and the US29).
Another international comparison was made between two epidemiological studies, the Multiethnic Study of Atherosclerosis (MESA) in the US and the Shiga Epidemiological Study of Subclinical Atherosclerosis (SESSA) in Japan, conducted among American and Japanese men aged 45–74 years33). This comparison showed that White men in the US had a higher prevalence of coronary atherosclerosis detected by CAC than Japanese men, after considering traditional coronary risk factors. However, the racial and ethnic differences were smaller in younger age groups (e.g., adjusted odds ratios for White men with a CAC score of ≥ 100 were 2.05, 2.43, and 3.86 among those aged 45–54, 55–64, and 65–74 years, respectively)33). The progression of CAC was also prospectively compared among White, Black, Hispanic, and Chinese men in the US in MESA and among Japanese men in SESSA34). CAC progression was defined as CAC ＞0 in those without baseline CAC or the annualized change in CAC. Of note, Japanese men were significantly older (63 vs. 59 years), had higher measured blood pressure and cholesterol values, and had a higher prevalence of diabetes mellitus and smoking than the above four racial and ethnic groups in the US; however, men in the US were more likely than Japanese men to be obese and take statins. Despite an overall higher cardiovascular risk factor burden and age, Japanese men had a lower baseline CAC score than all four US racial and ethnic groups (median Agatston score, 40 in Japanese men vs. range 60–126 in US racial and ethnic groups). Adjusted for traditional risk factors, time between scans, and medication use, Japanese men with baseline CAC had significantly lower annualized CAC progression than US men (Fig.2), regardless of race and ethnicity, and lower rates of conversion from CAC=0 than US White men. These results suggest a higher coronary atherosclerosis burden and consequent risk for CHD among US men than among Japanese men.
CAC progression was prospectively compared between 455 White, 211 Black, 183 Hispanic, and 109 Chinese men in the US and 464 Japanese men in Japan, aged 45–74 years, with detectable CAC (CAC score ＞0) at baseline.
Robust regression was used for the annual change in CAC, calculated as CAC score in follow-up minus CAC score at baseline divided by time between CT scans in years (median, 3.4 years in MESA and 5.2 years in SESSA).
Adjusted for CT scanner pairs; age, education, cigarette smoking status, pack-years of smoking, body mass index, systolic blood pressure, antihypertensive medication use, total cholesterol, high-density lipoprotein cholesterol, statin use, and diabetes mellitus at baseline; and changes in cigarette smoking status, pack-years of smoking, body mass index, systolic blood pressure, antihypertensive medication use, total cholesterol, high-density lipoprotein cholesterol, statin use, and diabetes mellitus between CT scans.
CI, confidence interval. ＊P＜0.01; †P＜0.001.
A discrepancy in the duration of exposure to coronary risk factors may be due to the difference in CAC burden between the US and Japan. In fact, even if current total cholesterol levels between the two populations are comparable, the cumulative association of total cholesterol with atherosclerosis may be smaller among Japanese men than among US men, especially for older individuals35). The likely hypothesis is also that there is a common source of exposure among Japanese men in Japan. The large difference in diet appears to be the strongest candidate that can explain the difference in CAC between the US and Japan, with consumption of fish and soy showing the largest differences. Researchers in the ERA-JUMP study found that (1) serum levels of marine n−3 fatty acids are more than 100% higher in Japanese than in US White and Japanese American men; (2) in each population, serum marine n−3 fatty acids had an inverse association with subclinical atherosclerosis, such as CAC and carotid intima–media thickness; and (3) higher levels of marine n−3 fatty acids among Japanese men in Japan significantly contributed to differences in the levels of subclinical atherosclerosis between Japanese men in Japan and White men in the US32, 36).
Similar to CHD, the CAC score is also related to traditional cardiovascular risk factors in Western as well as Asian populations37, 38). In the MESA and SESSA comparisons, regardless of country, race, or ethnicity, higher CAC scores were more common in men who had worse traditional cardiovascular risk factor profiles, such as hypertension, dyslipidemia, diabetes, smoking, and obesity33).
Epidemiological Study of CAC in Japan: the SESSAEpidemiological studies investigating subclinical atherosclerosis, including CAC, offer the advantage of a detailed examination of risk factors and biomarkers associated with asymptomatic atherosclerotic burden before the onset of ASCVD, despite a relatively small sample size and relatively short follow-up. The SESSA is one of the first epidemiological studies in Japan to assess measures of subclinical atherosclerosis in multiple vascular beds in a population-based cohort. Using SESSA data that started in 2006, we examined the prevalence and progression of CAC using the Agatston method on serial CT scans and characterized their cross-sectional or prospective associations with conventional and novel risk factors. For example, among conventional risk factors, cigarette smoking was strongly associated with a higher burden of CAC (Fig.3)39). Additionally, dose–response relationships of pack-years and daily cigarette consumption for CAC burden were observed among both current and former smokers. Even a small number of pack-years or low daily consumption among current smokers was associated with CAC, whereas the time since cessation among former smokers was linearly associated with a lower burden of CAC (Fig.3)39). We have also reported a series of novel ASCVD risk factors, including insulin resistance40), marine omega-3 fatty acids (serum eicosapentaenoic acid and docosahexaenoic acid)41), myokine (serum irisin)42), blood pressure variability43), micronutrients (serum magnesium, phosphorus, and calcium)44), lipoprotein particle profiles45, 46), serum LOX-1 ligand containing apolipoprotein AI47), serum proprotein convertase subtilisin/kexin type 9 48), and equol-producing status49) in relation to the presence or progression of CAC or other types of subclinical atherosclerosis.
Bar graphs indicate odds ratios for prevalent CAC, defined as Agatston CAC score ＞0, adjusted for age, body mass index, alcohol intake, systolic blood pressure, diabetes mellitus, total cholesterol, high-density lipoprotein cholesterol, antihypertensive medication use, lipid-lowering therapy, exercise habits, and C-reactive protein.
Error bars indicate 95% CIs.
Divided according to tertiles of pack-years of smoking (T1, ＜29.1; T2, 29.1–46.7; T3, ≥ 46.8) or time since cessation (T1, ＜10.4 years; T2, 10.4–24.3 years; T3, ≥ 24.4 years).
＊P＜0.05; †P＜0.01.
CI, confidence interval.
The effect of alcohol consumption on CHD is still unclear50). Observational studies have demonstrated that, compared with no consumption, light to moderate alcohol consumption is associated with a reduced risk of clinical CHD; there is less evidence of a protective effect on subclinical CHD as measured using CAC. Given the inability to conduct a randomized trial to confirm or refute the effect of alcohol consumption on CHD, an alternative approach is to use a genetic variant as a proxy for alcohol consumption. The single-nucleotide polymorphism (SNP) rs671 in the aldehyde dehydrogenase 2 gene (ALDH2), which encodes the ALDH2 enzyme, provides the primary pathway of alcohol metabolism. The ALDH2-rs671 variant, common only in East Asian populations, greatly slows acetaldehyde breakdown, and the resulting accumulation of acetaldehyde can cause severe discomfort that strongly reduces alcohol consumption. This can lead to the selection of the SNP rs671 as a genetic instrument in Mendelian randomization analysis. The aim of our paper from SESSA51) was therefore to investigate the causal role of alcohol intake in relation to subclinical and clinical CHD in two Mendelian randomization studies using the ALDH2-rs671 variant. The first investigation was a cross-sectional study of CAC using CT of 1,029 apparently healthy men, and the second was a case–control study of 421 men with CHD (acute coronary syndrome or stable angina pectoris) who underwent coronary revascularization and 842 age-matched male controls. Results of these analyses indicated a positive association of alcohol consumption with CAC burden but an inverse association of alcohol consumption with clinical CHD, especially acute coronary syndrome. These findings provide new hypothetical insight into how alcohol may reduce the risk of clinical CHD by introducing alterations in the plaque composition in coronary arteries from lipid to fibrous tissue with calcification, thereby decreasing the vulnerability of plaque to rupture and decreasing the risk of clinical CHD52).
Epidemiological Studies of CAC in Western CountriesTraditional risk factors such as age, sex, blood pressure, lipids, diabetes, and smoking, as supported by evidence, are the main focus of current guidelines for the primary prevention of ASCVD in Western countries and Japan. By incorporating these traditional risk factors into prediction models, the absolute risk of ASCVD events in individuals can be calculated. Some examples of prediction models include the Pooled Cohort Equations for ASCVD Risk Prediction in the US53, 54), SCORE in Europe55, 56), NIPPON DATA80 Risk Chart57), and Suita58), Hisayama59), and JALS60) SCORE in Japan. The concept of absolute risk, which considers the probability of future ASCVD events in individual patients, has become increasingly important in clinical practice. Understanding absolute risk is useful for the comprehensive management of risk factors, making decisions regarding medication, and other clinical interventions.
In recent years, there has been a growing effort to improve accuracy in predicting the onset of ASCVD by incorporating emerging biomarkers in addition to traditional risk factors. Among various biomarkers, the CAC score in particular has been gaining attention in Western countries as a highly predictive marker of ASCVD61). A large number of epidemiological studies using population-based cohorts have found that the total amount of CAC independently predicts future ASCVD, including CHD, and provides predictive information beyond what standard cardiovascular risk factors offer (Table 1). For example, in 2008, Detrano et al.62) published a groundbreaking paper from the MESA that explored the relationship between CAC and outcomes of ASCVD. The study followed participants for a median of 3.8 years and found that the adjusted risk of CHD events was 7.7-fold higher among individuals with CAC scores between 101 and 300 and 9.7-fold higher in those with scores above 300 than with individuals who had a score of 0. Importantly, their study was the first report of consistency in risk prediction across four main racial and ethnic groups (i.e., White, Black, Hispanic, and Chinese men in the US) with improvement in the c-statistic beyond traditional risk factors. More recently, Budoff et al.63) conducted a study that further supported the prognostic value of CAC in MESA. Their 10-year follow-up study showed that ASCVD event rates widely varied depending on age, sex, and racial and ethnic subgroups, with rates ranging from 1.3% to 24.5% based on the CAC score. The study also found that individuals with CAC scores above 100, regardless of demographic factors, had a 10-year ASCVD risk greater than 7.5%, indicating eligibility for statin therapy according to guidelines. In addition to earlier studies from MESA, Polonsky et al.64) conducted a study that assessed the ability of CAC to improve risk reclassification. They found that the addition of CAC to a model of age, sex, and risk factors resulted in a 25% net reclassification improvement; an additional 23% of those who experienced events were reclassified as high risk, and an additional 13% without events were reclassified as low risk when CAC was added. Furthermore, Yeboah et al.65) demonstrated that adding CAC to the Framingham risk score significantly improves the c-statistic (from 0.623 to 0.784, P＜0.001) compared with other risk markers such as the carotid intima–media thickness, flow-mediated dilatation, C-reactive protein, family history, or ankle–brachial index. These findings support the American College of Cardiology (ACC)/American Heart Association (AHA) 2013 risk assessment guideline66), which suggests that assessing CAC is the most useful approach to improving risk assessment in individuals found to be at intermediate risk after formal risk assessment. The MESA also investigated the prognostic importance of CAC progression in 5,682 adults who underwent repeat CAC scans after an average of 2.5 years, with a median follow-up for CHD events of 7.7 years67). The results showed that the cumulative incidence of CHD events varied depending on the extent of CAC progression. Specifically, the incidence ranged from less than 10% in individuals with annual CAC score increases of less than 100 units to over 30% in those with annual CAC score increases exceeding 300. In the Coronary Artery Risk Development in Young Adults (CARDIA) study of 3,043 adults aged 32–46 years who were scanned for CAC and followed for 12.5 years, Carr et al.68) found that young adults with any CAC have a 5-fold greater risk of CHD events and a 3-fold greater risk of ASCVD events. Additionally, within CAC score strata of 1–19, 20–99, and 100 or greater, the hazard ratios for CHD events were 2.6 (95% confidence interval [CI] 1.0–5.7), 5.8 (95% CI 2.6–12.1), and 9.8 (95% CI 4.5–20.5), respectively. In the last few decades, significant sex-specific differences in the epidemiology of ASCVD have been studied and established. In other words, women develop ASCVD approximately 10 years later in life than men69). In the CAC Consortium70), across age groups, women had a lower CAC prevalence than men, and within CAC subgroups, women had fewer calcified lesions and a lower CAC volume. ASCVD mortality among women and men with a CAC score of 0 was similar. Yet, a disproportionately higher ASCVD mortality (1.3-fold higher) was observed in women than in men when CAC was present (CAC score ＞0).
| Study | Country | Year of CAC measurement | No. of participants | Age, years | Years of follow-up | Association of CAC with outcomes＊ | Incremental values of CAC† |
|---|---|---|---|---|---|---|---|
| Rotterdam study89) | Netherlands | 1997-2001 | 3678 | Range, 55 or older | Median, 6.8 | Adjusted HR (95% CI) for hard CHD events, 6.2 (3.4–11.5) for 4th vs. 1st quartile of the CAC burden |
-C-statistic: traditional risk factors ＋CAC score, ＋0.05 (0.02–0.06) -NRI for Major CHD events: in entire study cohort, 0.19 (0.13– 0.26); in participants at intermediate risk (FRS), 0.39 (0.27–0.52) |
| MESA62) | USA | 2000-2002 | 6722 |
Range, 45–84 Mean, 62.2 |
Median, 3.8 |
Adjusted HR (95% CI) for any CHD events: CAC score of - 1–100, 3.6 (2.0–6.7) - 101–300, 7.7 (4.1–14.5) - ＞300, 9.7 (5.2–18.0) |
-AUC for any CHD events: risk factors alone, 0.77; risk factors＋ CAC score, 0.82, P＜0.001 -NRI for Any CHD events: in entire study cohort, 0.25 (0.16– 0.34), P＜0.001; in participants at intermediate risk (FRS), 0.55 (0.41–0.69), P＜0.001 64) |
|
Dallas Heart Study90) |
USA | 2000-2002 | 2084 | Mean, 44.4 | Mean, 9.2 |
Adjusted HR (95% CI) for any CHD events: CAC score of - 1–10, 1.0 (0.4–2.7) - 11–100, 3.4 (1.4–8.6) - ＞100, 5.6 (2.3–14.0) |
-C-statistic for any CHD events: risk factors alone, 0.86 (0.83– 0.91); risk factors＋CAC score, 0.89 (0.86–0.93), P= 0.003 -NRI in entire study cohort: for any CHD event, 0.22, P= 0.012; for hard CHD events, 0.31, P= 0.014 |
| Heinz Nixdorf Recall study91) | Germany | 2000-2003 | 4129 | Range, 45–75 | Median, 5 |
Adjusted HR (95%) for hard CHD events: CAC score of - 1–99, 1.4 (0.7–2.9) - 100–399, 2.8 (1.3–6.0) - ≥ 400, 6.4 (3.1–13.1) |
-AUC for hard CHD events: FRS alone, 0.68; FRS＋CAC score, 0.75, P= 0.003 -NRI for hard CHD events: for intermediate risk of 10–20%, 0.22, P＜0.001; for intermediate risk of 6–20%, 0.20, P= 0.004 |
| Framingham Hart Study92) | USA | 2002-2005 | 3486 | Mean, 50 | Median, 8 |
Adjusted HR (95%) for major CHD events: CAC score of - 1–100, 1.5 (0.6–3.8) - 101–300, 4.6 (1.7–12.4) - ＞300, 9.4 (3.6–24.4) |
-C statistic for major CHD events: Framingham risk factors alone, 0.78; Framingham risk factors + CAC score, 0.82; Framingham risk factors＋CAC category, 0.83 -NRI for major CHD events in entire study cohort: for CAC score, 0.32 (0.11–0.53); for CAC category, 0.22 (0.01–0.42) |
|
CARDIA study68) |
USA | 2000-2001 | 3036 | Mean, 40.3 | Mean, 12.5 |
Adjusted HR (95%) for all CHD events: CAC score of - 1–19: 2.6 (1.0–5.7) - 20–99: 5.8 (2.6–12.1) - ≥ 100: 9.8 (4.5–20.5) |
C statistic and NRI were not assessed. |
＊In all studies, the reference category was defined as CAC score = 0 in regression models.
†Net reclassification improvement (NRI) in intermediate-risk groups is noted to have important statistical limitations; however, the pre-eminence
of CAC compared with other tests has also been shown using other statistics.
Abbreviations: CAC, coronary artery calcium; ASCVD, atherosclerotic cardiovascular disease; HR, hazard ratio; USA, United States; CHD, coronary heart disease; AUC, area under the receiver operating characteristic curve; CI, confidence interval; MESA, Multi-Ethnic Study of Atherosclerosis; FRS, Framingham risk score; CARDIA, Coronary Artery Risk Development in Young Adults.
The effect of CAC on future stroke events was previously evaluated in four population-based studies: the Cardiovascular Health study71), the Rotterdam study72), MESA73), and the Heinz Nixdorf Recall study74). Notably, in the former two studies, CAC was not found to represent a stroke predictor when adjustments for traditional risk factors were made due to the small number of events. Based on the results from the Heinz Nixdorf Recall study and MESA, the CAC score is an independent factor for predicting stroke events after adjusting for conventional risk factors, regardless of stroke subtypes73, 74). In contrast, the CAC score alone may have little power for risk discrimination in incident stroke events. The CAC score might have favorable discriminative value for stroke in individuals with low to intermediate risk, but additional investigations restricted to only that population are needed to confirm the value of the CAC score75).
CAC for Primary Prevention of ASCVDACC/AHA recommendations include consideration of risk-enhancing factors to guide clinician–patient risk discussions for adults with intermediate risk (7.5%–20% 10-year ASCVD risk) and adults with borderline risk (5%–7.5% 10-year ASCVD risk)76, 77). These include patients with a family history of premature ASCVD, persistently high levels of LDL cholesterol ≥ 160 mg/dL or triglycerides ≥ 175 mg/dL, chronic kidney disease, metabolic syndrome, conditions specific to women (e.g., preeclampsia, premature menopause), inflammatory diseases (e.g., rheumatoid arthritis, psoriasis, and HIV), high-risk race and ethnicity (e.g., South Asian origin), and elevated levels of biomarkers (e.g., high-sensitivity C-reactive protein or lipoprotein(a)). In cases where the decision regarding preventive interventions remains unclear, it is recommended to consider CAC as an adjudicator to upgrade risk (e.g., young patients and women) or downgrade risk (e.g., elderly people, patients with diabetes)78). In 2013, the ACC/AHA ASCVD risk management guidelines substantially broadened the eligibility criteria for statin therapy in the adult population66). In this setting, CAC would perform robustly as a decision aid. The 2018 AHA/ACC/Multi-Society Cholesterol Guidelines77) and 2019 ACC/AHA Primary Prevention Guidelines76) both endorse the use of CAC scoring as a tool for shared decision-making in personalized risk management for primary prevention. Compared with the 2013 guidelines66), the 2018/2019 recommendations not only expanded the estimated 10-year ASCVD risk range within which CAC could be considered for personalized risk stratification (from 5%–7.5% to 5%–20%), but most CAC-related recommendations were also upgraded to IIa. Additionally, the recommendations provide detailed guidance on the use of CAC=0 scores to identify individuals who may not benefit from statin therapy if they are not at high risk or do not have diabetes, a family history of premature CHD, or a history of cigarette smoking.
The recommendations of the US Preventive Services Task Force (USPSTF) conflict with those of other organizations, such as the ACC and AHA, which recommend consideration of CAC testing in certain populations for ASCVD prevention79). The USPSTF’s 2018 statement concludes that there is insufficient evidence to support the addition of CAC testing to traditional cardiovascular risk assessment in asymptomatic adults for ASCVD prevention. Three randomized clinical trials have evaluated the potential clinical impact of CAC scanning in asymptomatic adults. One trial involving 450 participants found no significant improvements in modifiable cardiovascular risk factors after 1 year80). Another trial involving 1,934 participants found decreases in risk factors after 4 years81). A third trial involving 56 postmenopausal women found that risk factors declined less in the CAC-screened group after 1 year82). No trials have found a difference in actual ASCVD events based on the results of CAC scanning79). Several potential harms are associated with CAC scanning. The first is exposure to radiation, which is estimated to be 1.7 mSV for a CAC scan using American College of Radiology registry data. This is 17-fold higher than that from a two-view chest radiograph and 4.5-fold higher than that from a mammogram83) (range, 0.4–2.1 mSV)79). The second potential harm is adverse psychological effects from patients being told that their CAC scan results are positive84). The third is the potential for CAC scan results to lead to subsequent unnecessary cardiac tests and interventions, or even noncardiac tests and interventions, owing to incidental findings such as lung nodules in individuals who are otherwise clinically stable85, 86). These additional tests and treatments come with associated costs for both patients and the health care system87).
Guidance on the use of CAC for ASCVD risk assessment from local scientific societies is limited in many Asian countries. The Japanese Atherosclerotic Society’s Prevention of Atherosclerotic Cardiovascular Diseases 2022 Guidelines conclude that there is limited scientific evidence to support the use of CAC screening to improve the prediction of ASCVD risk in primary prevention among Japanese individuals. In South Korea, CAC and CT angiography are currently widely used for clinical plaque screening, typically as part of health checkups for workers88). Whether aggressive use of CAC or CT angiography for screening in populations at low risk of CHD yields improved clinical outcomes remains unclear.
CAC is widely available, has been extensively studied, and is a highly specific marker of subclinical atherosclerosis. The CAC score is independently associated with incident CHD or ASCVD events and provides improved predictive values for risk estimation of ASCVD, including CHD, beyond traditional risk factors, more so than other measures of subclinical atherosclerosis. CAC testing facilitates the up- or downgrading of risk in asymptomatic patients with borderline or intermediate risk who seek more definitive risk assessment as part of a clinician–patient discussion, and CAC provides a model for initiating or intensifying preventive statin pharmacotherapies. Uniting CAC risk stratification with cholesterol-modifying treatment may serve as a model for individualizing primary ASCVD prevention and shared clinician–patient decision-making. However, whether the same benefits are present in other populations, such as the Japanese population, which has a lower risk of CHD than Western populations, remains uncertain; this question warrants further investigation. Additionally, randomized controlled trials are needed to demonstrate the usefulness and safety of CAC screening for assessing ASCVD in primary prevention. Together, the available evidence suggests that CAC will remain a key tool in personalized primary ASCVD preventive care in the coming years.
We thank Analisa Avila, MPH, ELS, of Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.
None.
