Associations of Cigarette Smoking with Coronary Artery and Extra-Coronary Calcification in Older Adults: The Atherosclerosis Risk in Communities Study

Abstract

Background and Aims:

Smoking is a major risk factor for atherosclerosis, yet its impact on coronary artery calcium (CAC) and extra-coronary calcification (ECC), which commonly manifests at older age, has not been fully characterized. We aimed to evaluate the relationships between smoking (status, pack-years, and years since cessation) and CAC and ECC among older adults.

Methods:

In 1,901 Atherosclerosis Risk In Communities (ARIC) Study (United States) participants (mean age 81 ± 4 years, 62% female, 20% Black) without a clinical history of coronary heart disease at visit 7 (2018–19), we quantified the associations of key smoking parameters (status, pack-years, and years since cessation) with CAC and ECC (thoracic aorta and aortic/mitral valve calcification) using multivariable logistic regression models adjusted for other risk factors. Smoking parameters were lagged by 5 years. High CAC and ECC were defined as sex-race specific >75th percentile of Agatston score.

Results:

Current and former smoking were significantly associated with high CAC (adjusted odds ratio [aOR] 2.57 [95%CI 1.58–4.19] and 1.66 [1.30–2.12], respectively) and ECC except for the aortic valve. Pack-years demonstrated a robust dose-response relationship with CAC and ECC, but less so for the aortic and mitral valves. The association was particularly strong for calcification at aortic valve ring, ascending aorta, and descending aorta (e.g., aOR 4.10 [2.57– 6.55], 4.07 [2.56, 6.46], and 3.97 [2.45, 6.43], respectively, for current vs. never smokers). The odds of high vascular calcification declined with progressively longer years of smoking cessation.

Conclusions:

Cigarette smoking was strongly associated with vascular calcification, especially calcification in the aorta, but less so for valvular calcification. These findings reinforce the harm of smoking on broad ranges of vascular beds and simultaneously support unique pathophysiology across different vascular beds and cardiac valves.

Graphical Abstract

Introduction

Cigarette smoking (henceforth smoking) is a major risk factor for atherosclerosis[1], but its contribution may vary across different vascular beds. For example, several studies[2,3] have demonstrated that smoking is a stronger risk factor for peripheral artery disease (PAD) than coronary heart disease (CHD) or stroke. A Mendelian randomization study[4] has confirmed stronger causality of smoking for PAD than other atherosclerotic diseases.

Coronary artery calcium (CAC) measured by computed tomography (CT) is a strong indicator of coronary atherosclerosis and is an established predictor of future cardiovascular events.[5] Moreover, accumulating evidence suggests additional prognostic value beyond CAC of extra-coronary calcification (ECC) (such as calcification in the thoracic aorta and cardiac valves).[6] However, the associations of different parameters of cigarette smoking with calcification of different vascular beds and cardiac valves have not been fully characterized.

Specifically, a number of studies have shown a consistent dose-response association between smoking and CAC;[7–15] however, inconsistent results were reported for thoracic aortic calcification[7,13,16–20] and aortic valve calcification.[7,20–25] Importantly, few studies have directly compared the contribution of smoking to calcification of different vascular beds and cardiac valves in the same study population.[7] Also, most studies used a simple categorization of smoking status (current, former, or never), and it is inconclusive whether and how long the impact of smoking on vascular calcification across different vascular beds lasts after smoking cessation.[15,16]

Therefore, we sought to comprehensively quantify the associations of cigarette smoking (smoking status and pack-years) and smoking cessation with calcification of different vascular beds and cardiac valves in older adults, using data from a community-based cohort study, the Atherosclerosis Risk in Communities (ARIC) Study.

Methods

Study Population

The ARIC study is a community-based cohort that enrolled 15,792 participants—primarily White and Black individuals aged 45 to 64 years—during 1987 and 1989 (visit 1) from four US communities: Washington County, MD; suburban Minneapolis, MN; Forsyth County, NC; Jackson, MS. Institutional Review Boards of all study centers approved the study, which complied with the Declaration of Helsinki and secured written informed consent from every participant.

At visit 7 (2018–2019), participants without a clinical history of CHD were invited to undergo a non-contrast cardiac-gated CT scan for the first time in ARIC since data on prognostic value of CAC in the 75-and-older population were sparse. Among 3,589 participants at visit 7, 3,126 participants free of CHD before 2016 were eligible for a CT scan. Of those, 2,290 participants (73.3%) consented to the investigation. The final study dataset included 1,901 participants from visit 7 after excluding those with missing smoking data (n = 224), missing data on covariates (n = 331), prevalent CHD or stroke before CT scan (n = 585), missing calcification data on any vascular beds (n = 540), and individuals self-identifying as races others than Black or White people due to small sample size (n = 8) (Supplemental Figure 1).

Exposure: Cigarettes Smoking

We used the smoking parameters from ARIC visit 1 to ARIC visit 5 (2011–13) to have a lag time of ~5 years of smoking exposure before CAC/ECC measurement at visit 7 (2018–19) since the development and progression of vascular calcification are usually not rapid processes.[26,27] Smoking status (current, former, and never smoker) at visit 5 was obtained through a self-reported questionnaire. For current or former smokers, pack-years were calculated by incorporating the pack-years at visit 1,[3] along with smoking status and intensity (pack per day) from the subsequent visits 2 through 5. For instance, if a participant who was a current or former smoker at visit 1 answered “yes” to “Do you now smoke cigarettes?” at visit 2, additional pack-years were added based on the duration between visits 1 and 2 and smoking intensity at visit 1 (duration in years multiplied by packs per day), summed until the participant reported cessation.

Among former smokers at visit 5, the years since cessation at visit 5 was assessed using information at visit 1 and updated cessation data from each subsequent study visit or telephone interview. For example, if a former smoker mentioned quitting 10 years by visit 1 and responded “no” to “Do you now smoke cigarettes?” at all subsequent visits, the intervals before visit 5 (e.g., 35 years) were accumulated into years since smoking cessation. If a participant identified as a current smoker at visit 2 responded “no” to the same question at visit 3, the start of cessation was set at the midpoint between the date of visit 2 and 3. For participants who were “on and off” smokers, we only counted the quitting years after the most recent data point when they were current smokers (e.g., if a participant reported current smoking at visits 1 and 3 but not smoking at visits 2 and 4, the cessation initiation was set at the midpoint between visits 3 and 4).

Outcome variables: CAC and ECC

At each center, CT scans were obtained using 64-slice multidetector CT scanners operated by trained technicians under physician supervision. Calcium amount was quantified by the Agatston scoring method.[28] Calcification was defined as any lesion per slice exhibiting an attenuation of ≥130 Hounsfield Units (HU) and an area of ≥1 mm². The area of each calcified lesion was multiplied by a density factor based on the lesion’s highest HU (130–199 HU =1, 200–299 HU =2, 300–399 HU =3, and ≥400 HU =4). The scores were summed for all lesions across all slices.

CAC was defined as calcification in any of the coronary arteries. ECC was defined as calcification at five specific sites: ascending aorta, descending thoracic aorta, aortic valve ring (the aorta segment below the coronary arteries and above the aortic valve), aortic valve, and mitral valve. High CAC and ECC were defined as >75th percentile of the sex- and race-specific Agatston score in our ARIC study population for each calcification site.[29]

Covariates

We used covariate data at visit 7 unless specified otherwise. Age at visit 7 was calculated by subtracting the birthdate from the date of visit 7. Sex, race, and education level were recorded at visit 1; education was categorized as basic (less than completed high school), intermediate (high school graduate or vocational school), and advanced (at least some college). Alcohol consumption status was self-reported as current, former, and never drinker. Body mass index (BMI) was determined by dividing weight in kilograms by height in meters squared (kg/m²). Systolic and diastolic blood pressures were measured three times using a sphygmomanometer after a 5-minute rest, and the average of the last two readings was recorded. Plasma total cholesterol and high-density lipoprotein (HDL) cholesterol levels were measured using automated enzymatic methods.[30] Estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) creatinine equation.[31] Medication use in the past four weeks was self-reported and verified by inspecting medication containers brought to the study site. Hypertension was defined as systolic blood pressure (SBP) ≥140 mmHg, diastolic blood pressure (DBP) ≥90 mmHg, or the use of antihypertensive medications. Diabetes mellitus was identified by fasting glucose ≥126 mg/dL, non-fasting glucose ≥200 mg/dL, self-reported physician diagnosis, or use of anti-diabetic medication.

Statistical Analysis

Participant characteristics were summarized according to smoking status (current, former, never) using chi-square tests for categorical variables and analysis of variance (ANOVA) for continuous variables.

To investigate the association of smoking parameters (smoking status, pack-years, and years since cessation) with high CAC and ECC, we ran multivariable logistic regression models and estimated odds ratios (ORs). Unless specified otherwise, never smokers served as the reference group in all analyses. We considered two models to account for potential confounders: Model 1 adjusted for age, sex, race, and study center; Model 2 further adjusted for education level, BMI, SBP, total cholesterol, HDL cholesterol, eGFR, alcohol consumption status (current, former, never), diabetes, antihypertensive medication use, and cholesterol-lowering medication use. Pack-years were analyzed both as categorical variables (<20, 20– <40, and ≥40 pack-years) and continuous variables (per 10 pack-years or in cubic spline terms with knots at 10, 20, 40, and 60 pack-years). Years since cessation was categorized as <10, 10– <30, and ≥30 years since cessation.

Several sensitivity analyses were also performed. First, we repeated the analyses using the presence of any calcium (Agatston score >0 vs. 0) at different sites as the outcome variable. Second, we conducted subgroup analyses stratified by pre-specified groups: age (> vs. ≤ 81 years [median]), sex (female vs. male), race (Black vs. White people), BMI (> vs. ≤ 28 kg/m² [median BMI]), and the presence vs. absence of hypertension, and diabetes. In these analyses, former and current smokers were combined to obtain more reliable estimates. The likelihood ratio test was utilized to assess potential effect modification, comparing models (including all covariates from Model 2) with and without an interaction term of interest.

All analysis was conducted using Stata version 16.0 (StataCorp LP, College Station, TX, USA), and a p-value <0.05 was considered statistically significant.

Results

Baseline Characteristics

In this study population, mean age was 81 (standard deviation [SD]: 4) years; 63% were female, and 20% were Black persons (Table 1). There were 1,000 (53%) never smokers, 807 (43%) former smokers, and 94 (4%) current smokers. Compared to never smokers, current smokers were generally younger and more likely to be male and Black persons, current drinkers, have less years of education and exhibit lower SBP, total cholesterol, and eGFR. They also tended to have a lower prevalence of diabetes and use of cholesterol-lowering medication. Generally, former smokers displayed characteristics intermediate between never and current smokers, except they had the lowest proportion of current drinkers, the highest BMI, and the lowest SBP.

Table 1.

Baseline Characteristics by Smoking Status, the ARIC Study (N=1901).

	Overall	Never Smokers	Former Smokers	Current Smokers
No. of participants	1901	1000 (52.6%)	807 (42.5%)	94 (4.9%)
Age, years	80.9 (4.2)	81.1 (4.3)	81.0 (4.2)	78.9 (3.2)
Female	1186 (62.4%)	709 (70.9%)	425 (52.7%)	52 (55.3%)
Black	383 (20.2%)	220 (22.0%)	143 (17.7%)	20 (21.3%)
Study Center
Forsyth County, NC	416 (21.9%)	214 (21.4%)	176 (21.8%)	26 (27.7%)
Jackson, MS	344 (18.1%)	204 (20.4%)	125 (15.5%)	15 (16.0%)
Minneapolis, MN	661 (34.8%)	309 (30.9%)	319 (39.5%)	33 (35.1%)
Washington County, MD	480 (25.3%)	273 (27.3%)	187 (23.2%)	20 (21.3%)
Education Level at visit 1
Basic level	172 (9.1%)	94 (9.4%)	71 (8.8%)	7 (7.4%)
Intermediate	773 (40.7%)	405 (40.5%)	320 (39.7%)	48 (51.1%)
Advanced	956 (50.3%)	501 (50.1%)	416 (51.5%)	39 (41.5%)
BMI, kg/m2	28.0 (5.3)	28.0 (5.4)	28.2 (5.2)	27.0 (4.9)
Drinking status
Current	1024 (54.0%)	454 (45.4%)	514 (64.1%)	56 (59.6%)
Former	463 (24.4%)	214 (21.4%)	220 (27.4%)	29 (30.9%)
Never	408 (21.5%)	331 (33.1%)	68 (8.5%)	9 (9.6%)
Systolic blood pressure, mmHg	134.7 (18.9)	135.4 (18.9)	133.9 (18.7)	134.2 (19.9)
Diastolic blood pressure, mmHg	66.5 (10.7)	66.4 (10.5)	66.7 (10.9)	66.8 (12.1)
Total cholesterol, mg/dL	4.6 (4.0, 5.3)	4.7 (4.0, 5.4)	4.6 (4.0, 5.2)	4.4 (3.7, 5.2)
HDL cholesterol, mg/dL	1.3 (1.1, 1.6)	1.3 (1.1, 1.6)	1.3 (1.1, 1.5)	1.3 (1.1, 1.5)
eGFR based on serum creatinine, mL/min per 1.73 m2	62.0 (50.0, 73.0)	62.0 (50.0, 72.0)	62.0 (49.0, 73.0)	58.0 (49.0, 69.0)
Antihypertensive medication use	1239 (66.0%)	660 (66.7%)	520 (65.5%)	59 (63.4%)
Cholesterol-lowering medication use	959 (50.5%)	486 (48.6%)	418 (51.8%)	55 (58.5%)
Diabetes mellitus	530 (28.1%)	286 (28.8%)	219 (27.3%)	25 (26.6%)

Values for categorical variables are given as a number (percentage). Values for continuous variables are given as mean (standard deviation) and median (interquartile interval) if skewed distribution.

All p values from ANOVA for continuous variables and chi-square test for categorical variables by smoking status were <0.05.

ARIC = Atherosclerosis Risk in Communities; BMI = body mass index; eGFR = estimated glomerular filtration rate; HDL = high-density lipoprotein.

Association of Smoking with CAC and ECC

In demographically adjusted models (Model 1 in Table 2), compared with never smokers, current smokers demonstrated a significantly higher odds of high CAC and ECC in all vascular beds except that in aortic valve (borderline significant in mitral valve). For example, the adjusted odds ratio [aOR] was 2.54 [1.59–4.05] for the coronary arteries. Former smokers exhibited an intermediate odds of high calcifications between never and current smokers (e.g., aOR 1.60 [1.28–2.01] for the coronary arteries in Model 1). The patterns remained consistent after further adjustment for potential confounders (Model 2 in Table 2). The strongest associations were observed for high calcification at the aortic valve ring, ascending aorta, and descending aorta (aOR 4.10 [2.57–6.55], 4.07 [2.56–6.46], 3.97 [2.45–6.43], respectively, comparing current vs. never smokers in Model 2).

Table 2.

Adjusted Odds Ratio (95% CIs) of Coronary Artery Calcification and Extracoronary Calcification Scores (Agatston Score > vs. ≤ 75th Percentile) by Smoking Status.

		Smoking Status
		Never smoker (N=1000)	Former smoker (N=807)	Current smoker (N=94)
Total of coronary arteries	Model 1	Ref.	1.60 (1.28, 2.01) a	2.54 (1.59, 4.05) a
	Model 2	Ref.	1.66 (1.30, 2.12) a	2.57 (1.58, 4.19) a
Ascending aorta	Model 1	Ref.	1.39 (1.10, 1.74) a	4.05 (2.59, 6.32) a
	Model 2	Ref.	1.36 (1.06, 1.73) a	4.07 (2.56, 6.46) a
Descending aorta	Model 1	Ref.	1.60 (1.27, 2.02) a	3.96 (2.50, 6.29) a
	Model 2	Ref.	1.65 (1.29, 2.12) a	3.97 (2.45, 6.43) a
Aortic valve ring	Model 1	Ref.	1.40 (1.12, 1.75) a	3.75 (2.39, 5.88) a
	Model 2	Ref.	1.40 (1.10, 1.78) a	4.10 (2.57, 6.55) a
Aortic valve	Model 1	Ref.	1.30 (1.04, 1.62) a	1.21 (0.72, 2.04)
	Model 2	Ref.	1.30 (1.03, 1.65) a	1.26 (0.74, 2.14)
Mitral valve	Model 1	Ref.	1.11 (0.89, 1.38)	1.62 (1.00, 2.62) a
	Model 2	Ref.	1.10 (0.87, 1.40)	1.70 (1.03, 2.79) a

^ap value <0.05.

Model 1: adjusted for age, gender, race, study center; Model 2: adjusted for Model 1 covariates, plus education level, BMI, systolic blood pressure, total cholesterol, HDL cholesterol, eGFR, alcohol status (current, former, never), diabetes mellites, antihypertensive medication use, and cholesterol-lowering medication use.

CI = confidence interval; other abbreviations as in Table 1.

In sensitivity analysis using any calcification (>0 vs. 0 Agatston score) as the outcome variable (Supplemental Table 1), current smoking was significantly associated with high calcification in the aortic valve ring (aOR 4.24 [1.67, 10.79] vs. never smoking in Model 2), ascending aorta, and mitral valve, but no significant association was found for other vascular beds. Subgroup analysis (Supplemental Figure 2) demonstrated consistent associations of smoking status with CAC and ECC across various demographic and clinical subgroups with no significant interactions detected (all p for interaction >0.05).

We observed a dose-responsive relationship between greater pack-years and high CAC and ECC at most sites when we modeled pack-years continuously with cubic spline terms (Figure 1). Notably, the graded association plateaued after 20 pack-years for calcification in the coronary arteries, while the associations were linear for calcification in the aortic valve ring, ascending, and descending aorta. For calcification in the aortic or mitral valves, the association with pack-years was nearly flat. These patterns were confirmed when we modeled pack-years as a categorical variable (Supplemental Table 2). When we analyzed any calcification (>0 vs. 0 Agatston score) as the outcome variable, the patterns were largely consistent (Supplemental Table 3).

Figure 1. Adjusted Odds Ratios (95% CIs) for Agaston Score > vs. ≤ 75th Percentile of Calcification at Different Sites by Pack-years.

Solid lines indicate adjusted ORs and dashed lines indicate 95% CIs from logistic regression models using restricted cubic splines for pack-years (knots at 10, 20, 40, and 60 pack-years). Bars represent the histogram of pack-years in the analytic sample (proportion of participants in each pack-year bin; the bin of 0 pack-year corresponds to never smokers).

Model was adjusted for age, gender, race, study center, education level, BMI, systolic blood pressure, total cholesterol, HDL cholesterol, eGFR, alcohol status (current, former, never), diabetes, antihypertensive medication use, and cholesterol-lowering medication use.

Abbreviations as in Tables 1 and 2.

Association of Smoking Cessation with CAC and ECC

The odds of high calcification in most vascular beds was lower with longer years since cessation in generally a graded fashion (Figure 2), although even the category of current smoking or <10 years of cessation did not show significant associations with aortic or mitral valve calcification. For CAC, even the category of cessation >30 years showed a significant association with high calcification (aOR 1.55 [1.16–2.07] vs. never smoking).

Figure 2. Adjusted Odds Ratios (95% CIs) for Agaston score > vs. ≤ 75th Percentile of Calcification at Different Sites by Years of Cessation.

Model was adjusted for age, gender, race, study center, education level, BMI, systolic blood pressure, total cholesterol, HDL cholesterol, eGFR, alcohol status (current, former, never), diabetes, antihypertensive medication use, and cholesterol-lowering medication use.

Abbreviations as in Tables 1 and 2.

Discussion

In this community-based cohort of older adults with no history of clinical CHD, cigarette smoking showed a robust association with CAC and ECC. The association was particularly strong with calcification of the thoracic aorta (i.e., aortic valve ring, ascending aorta, and descending aorta) and consistent across subgroups by demographics and major atherosclerotic risk factors. There was a dose-responsive relationship between pack-years and calcification at different vascular beds except in cardiac valves. Moreover, the odds of high calcification in these vascular beds gradually decreased with longer years since cessation.

Our observations of a robust relationship between smoking and CAC (with ~2- to 2.5-fold higher odds of high CAC in former/current smokers vs. never smokers) were consistent with previous research.[7–15] Leveraging the comprehensive measurements of smoking parameters in the ARIC study, we were able to meticulously assess pack-years and smoking cessation over a few decades and their associations with CAC. A previous report[15] observed the decline of CAC risk with longer smoking cessation among Asian patients with chronic kidney disease, but we extended this finding to the US general populations of White and Black people. Notably, we found that the risk of high CAC persisted even after >30 years of smoking cessation.

Previous studies examining the association between smoking and ECC have inconsistent results and a few limitations, for example, often focusing solely on smoking status as the exposure,[7,17–25] or limiting their scope to either aortic calcification[13,16–20] or valvular calcification[21–25]. We found particularly evident associations between smoking and calcification in the thoracic aorta, stronger than the relationship observed with cardiac valvular calcification or even CAC. The differential impact of smoking on calcification across various vascular beds may be attributed to several potential mechanisms. Aortic calcification primarily involves the medial or both intimal and medial layers of the arterial wall, whereas CAC is mainly an intimal process associated with atherosclerosis.[32] The aorta's larger diameter and higher elastin content[33] may render it more susceptible to calcific changes induced by smoking-related oxidative stress and inflammation.[34,35] Nonetheless, our observation seems to be in line with the fact that smoking is a leading risk factor for aortic aneurysm.[36]

Our study carries important public health implications. The 2014 Surgeon General's report on the health consequences of smoking[37] expanded the list of CVDs (e.g., CHD, stroke, and PAD) caused by smoking to include abdominal aortic aneurysm (AAA), but given our findings, there is a compelling need to further our view of smoking's detrimental effects on the entire cardiovascular system, including the thoracic aorta. Importantly, thoracic aorta calcification has been a robust predictor of poor prognosis independently of CAC.[38,39] From another point of view, thoracic aorta calcification may serve as a marker of smoking-related vascular injury, reflecting both cumulative exposure and individual susceptibility, and may guide smoking-related clinical management. For example, there are a few screening guidelines targeting smokers, namely lung cancer using low-dose computed tomography (LDCT)[40] and AAA using ultrasonography,[41] but the adherence to those guidelines is suboptimal (e.g., lower than 50%).[42] Given our findings, those with eligible smoking history and thoracic calcification can be particularly encouraged to get those evidence-based screening procedures. Notably, thoracic aorta calcification can be identified by regular chest X-ray. Nonetheless, this concept requires future investigation to determine efficacy and cost-effectiveness. Additionally, we observed the persistence of increased calcification in former smokers, which further highlights the long-lasting vascular consequences of smoking. These insights underscore the importance of smoking prevention in youth.

The strengths of our study include the use of a well-conducted community-based cohort with detailed assessments of smoking behaviors over ~30 years and vascular calcification. The substantial representation of female and Black participants enhances the generalizability of our findings to these populations. However, several limitations should be acknowledged. First, vascular and valvular calcification was assessed at a single time point, precluding us from evaluating the incidence and progression of calcification. Second, the self-reported smoking measurements may be subject to recall bias and misclassification. However, smoking status and intensity were collected repeatedly across ARIC visits and smoking status was updated during annual telephone interviews between visits 4–5. Importantly, self-reported smoking status is generally reliable with a specificity and sensitivity of ~90%.[43] Third, non-contrast CT cannot capture non-calcified plaque. Nonetheless, previous studies reported a high correlation (r~0.90) between calcification volume and plaque volume.[44] Fourth, despite adjusting for major potential confounders, residual confounding by unmeasured factors cannot be entirely excluded. Finally, our study population was restricted to participants who survived and were able to undergo CT scanning at ARIC visit 7. As a result, heavier smokers and participants with more atherogenic risk factor profiles were less likely to be included in the present study, which may attenuate our associations of interest, although the direction and magnitude of bias are not certain. Also, our study consisted of older adults and predominantly comprised White and Black participants in the US, which may limit the generalizability of the results to younger individuals, other racial/ethnic groups, or non-US regions.

Conclusions

Cigarette smoking demonstrated a robust association with both CAC and ECC, with particularly strong associations with thoracic aorta calcification. Smoking cessation was associated with lower odds of calcification, but the elevated odds in certain vascular beds persisted for over 10 to 30 years. These findings further support the adverse impact of smoking on broad ranges of vascular beds and highlight distinct pathophysiological characteristics across various vascular beds and cardiac valves.

Highlights.

• Smoking is strongly and differentially associated with coronary and extra-coronary calcification in older adults.

• Current smokers have 2- to 4-fold higher odds of coronary and extra-coronary calcification, strongest in thoracic aorta.

• Pack-years show a clear dose-response with vascular calcification, especially in the thoracic aorta.

• Calcification risk declines with longer smoking cessation but can persist for decades.

• Thoracic aorta calcification may reflect cumulative exposure to smoking-related vascular injury and guide risk assessment.

Abbreviations:

CAC

coronary artery calcium

ECC

extra-coronary calcification

ARIC

Atherosclerosis Risk In Communities

CT

computed tomography

CAD

coronary heart disease

PAD

peripheral artery disease