The relation between coronary artery calcification in asymptomatic subjects and both traditional risk factors and living in the city centre: a DanRisk substudy


Jess Lambrechtsen, Department of Cardiology, Svendborg Hospital, DK-5700 Svendborg, Denmark. (fax: +45 63202225; e-mail:


Abstract.  Lambrechtsen J, Gerke O, Egstrup K, Sand NP, Nørgaard BL, Petersen H, Mickley H, Diederichsen ACP (Svendborg Hospital, Svendborg; Odense University Hospital, Odense; SVS, Esbjerg; Institute of Regional Health Services Research, University of Southern Denmark; Vejle Hospital, Vejle; and Odense University Hospital, Odense, Denmark). The relation between coronary artery calcification in asymptomatic subjects and both traditional risk factors and living in the city centre: a DanRisk substudy. J Intern Med 2012; 271: 444–450.

Objective.  To evaluate the association between the risk factor for living in the city centre as a surrogate for air pollution and the presence of coronary artery calcification (CAC) in a population of asymptomatic Danish subjects.

Design and subjects.  A random sample of 1825 men and women of either 50 or 60 years of age were invited to take part in a screening project designed to assess risk factors for cardiovascular disease (CVD). Noncontrast cardiac computed tomography was performed on all subjects, and their Agatston scores were calculated to evaluate the presence of subclinical coronary atherosclerosis. The relationship between CAC and several demographic and clinical parameters was evaluated using multivariate logistic regression.

Results.  A total of 1225 individuals participated in the study, of whom 250 (20%) were living in the centres of major Danish cities. Gender and age showed the greatest association with the presence of CAC: the odds ratio (OR) for men compared with women was 3.2 [95% confidence interval (CI) 2.5–4.2; < 0.0001], and the OR for subjects aged 60 versus those aged 50 years was 2.2 (95% CI 1.7–2.8; < 0.0001). Other variables independently associated with the presence of CAC were diabetes and smoking with ORs of 2.0 (95% CI 1.1–3.5; = 0.03) and 1.9 (95% CI 1.4–2.5, < 0.0001), respectively. The adjusted OR for subjects living in city centres compared to those living outside was 1.8 (95% CI 1.3–2.4; = 0.0003).

Conclusion.  Both conventional risk factors for CVD and living in a city centre are independently associated with the presence of CAC in asymptomatic middle-aged subjects.


The degree of cardiovascular disease (CVD) in an individual is affected by the lifetime exposure to various risk factors. For many years, CVD is asymptomatic, even in those with severe atherosclerosis. CVD is currently one of the most deadly and disabling diseases [1]. Subclinical coronary artery disease as determined by computed tomography (CT) imaging for the detection of coronary artery calcification (CAC) [2, 3] has been shown to provide powerful prognostic information beyond what is obtained from the analysis of traditional Framingham risk factors across a wide range of ages and ethnicities [4–7]. However, traditional risk factors constitute the most important preventive target [8]. In previous observational studies, long-term exposure to air pollution [9, 10] was related both to an individual’s place of residence and to the progression of CAC. Whether air pollution, socio-economic factors or lifestyle characteristics are the most significant risk factors associated with living in city centres is still a matter of debate [11]. To our knowledge, the influence of place of residence on the presence of CAC has not been previously studied. The purpose of this study was to evaluate the association between conventional risk factors for CVD as well as the place of residence and the presence of CAC in a cohort of asymptomatic middle-aged individuals.


The study design has previously been described in detail [12]. In brief, the cohort was a sample of 1825 men and women, 50 and 60 year old, randomly selected individuals from the Danish national Central Person Register in which all Danish citizens are registered. Patients with known CVD were not excluded in this first selection process. A total of 1257 subjects (69%) accepted the invitation to participate in the study. Before examination, the participants filled out a questionnaire concerning medical conditions (including previous CVD), current medication, smoking habits and family history of CVD. The screening study was performed in one of the four regional hospitals in the southern region of Denmark (in Odense, Esbjerg, Svendborg and Vejle). Heart valve disease was defined as self-reported routine follow-up with echocardiography or previous surgery for the condition.

The protocol was approved by the Central Ethical Committee and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from each participant.

Risk factors

Hypertension was defined as the use of antihypertensive medical treatment and/or blood pressure ≥140/90 mmHg at the time of examination. Diabetes mellitus was defined as the use of antidiabetic medication, fasting plasma blood glucose level ≥7.0 mmol L−1 on two different days or a postprandial blood glucose level ≥11.1 mmol L−1. Current medications were recorded. On the day of examination, weight, height and waist circumference were measured, and body mass index (BMI) was calculated. Blood pressure was recorded three times after resting for 5 min in the supine position; the last two values were averaged for the analyses. Blood samples were collected and laboratory tests performed to determine blood glucose and lipid parameters. Participants were not required to fast before all examinations.

A total of 251 participants living in the central areas of the five biggest cities of the region (31, 73, 65, 53 and 29 in Odense, Kolding, Fredericia, Vejle and Esbjerg, respectively) were selected by postcode and defined as ‘living in the city centre’. ‘Urban living’ was defined as living in the city but outside the city centre, and ‘rural living’ was defined as living outside the city.

Coronary calcification

Calculation of the total Agatston score was performed by summing the scores from each of the foci found in the coronary arteries [13]. To assess the Agatston score, 64-slice CT scanners were used and noncontrast CT scans were performed [12]. Two centres, Odense and Svendborg, used a GE 64-slice CT scanner (Discovery VCT; GE Healthcare, Milwaukee, WI, USA), and the scan was performed with the following parameters: gantry rotation time 500 ms, 16 × 2.5 mm collimation, 120 kV tube voltage, 200 mA tube current and a prospectively EGC-triggered scan acquisition gating at 50% of the R–R interval. In Vejle Hospital, a Siemens 64-slice Dual Source CT scanner (Siemens Definition; Siemens Medical Solutions, Erlangen, Germany) was used with the following parameters: gantry rotation time 330 ms, 3.0 mm collimation, 100–120 kV tube voltage, 150 mA tube current and prospective gating at 60% of the R–R interval. In Esbjerg Hospital, a Toshiba 64-slice CT scanner (Aquilion; Toshiba Medical Systems, Tokyo, Japan) was used with the following parameters: gantry rotation time 450 ms, 3 mm collimation, 120 kV tube voltage, 240 mA tube current and prospective gating at 75% of the R–R interval. In all cases, the scan data were acquired during an inspiratory breath hold. Experienced cardiologists in each of the four centres classified the Agatston score as zero (score 0 U), low (score 1–399 U) or high (score ≥400 U). Mean radiation dose differed between CT scanners with mean values of 0.9, 1.3 and 1.7 mSv for the GE Discovery VCT, Siemens Dual Source and Toshiba scanners, respectively (< 0.0001).

Air pollution

The air quality in Denmark is monitored from a network of measuring stations throughout the country, and the results are available online in annual reports [14]. Various pollutants are measured including particulate matter (PM) and the combination of nitrogen oxide and nitrogen dioxide (NOx). NOx level is related to traffic pollution intensity. Particle pollution from particles <10 μm in diameter (PM10) is also measured. These pollutants are measured in Odense city centre, and in suburban and rural areas throughout Denmark.


Data were analysed and displayed descriptively; continuous data were summarized by means of descriptive statistics (arithmetic mean and standard deviation) and categorical data as frequencies with corresponding percentages. Exploratory hypothesis testing on characteristics of the study cohort in relation to place of residence (city centre versus urban or rural; Table 1) was carried out using the Wilcoxon rank sum test for continuous variables and Fisher’s exact test (Fisher–Freeman–Halton test) for 2 × 2 (k × 2, k > 2) frequency tables.

Table 1.   Characteristics of the study cohort in relation to place of residence
 All participants (= 1225)City centre (= 251)Urban or rural (= 974)
  1. CVD, cardiovascular disease; BMI, body mass index.

  2. Values are n (%) or mean ± SD unless otherwise indicated.

50 years old603 (49%)120 (48%)483 (50%)
60 years old622 (51%)131 (52%)491 (50%)
Male580 (47%)121 (48%)459 (47%)
Female645 (53%)130 (52%)515 (53%)
Medical treatment
 Antihypertensive agents262 (21%)60 (24%)202 (21%)
 Lipid-lowering agents120 (10%)19 (8%)101 (10%)
Systolic blood pressure, mmHg136 ± 19128 ± 16138 ± 19
Diastolic blood pressure, mmHg82 ± 1081 ± 1082 ± 10
Cholesterol, mmol L−15.5 ± 1.05.6 ± 1.15.4 ± 1.0
LDL, mmol L−13.2 ± 0.93.5 ± 1.03.2 ± 0.9
 Current309 (25%)69 (28%)240 (25%)
 Former411 (34%)86 (34%)325 (33%)
 Non-smokers505 (41%)96 (38%)409 (42%)
Family history of CVD287 (23%)51 (20%)236 (24%)
BMI, kg m−227.0 ± 4.826.8 ± 4.727.0 ± 4.8
Waist circumference, cm95 ± 1294 ± 1295 ± 13
Agatston score, U (= 1221)
Mean, U87 ± 33993 ± 36185 ± 333
Median, U030
 0672 (55%)113 (45%)559 (57%)
 1–9153 (13%)43 (17%)110 (11%)
 10–99217 (18%)59 (24%)158 (16%)
 100–399116 (10%)20 (8%)95 (10%)
 >39963 (5%)14 (6%)49 (5%)

The association between CAC (Agatston score >0 versus Agatston score of 0) and various demographic as well as clinical variables was assessed by multivariate logistic regression. To define potential explanatory variables of CAC, univariate logistic regressions were applied on all demographic and clinical variables (i.e. sex, age, BMI, waist circumference, systolic and diastolic blood pressure, ankle–brachial index, resting heart rate, cholesterol, diabetes, smoking status, family history, haemoglobin A1c, LDL, HDL, triglycerides, C-reactive protein and living in the city centre). Statistically significant variables in the univariate analysis were included in a multivariate logistic regression model, in which stepwise subset selection was applied for adjustment. The significance level was 5%. All analyses were performed using sas 9.1.3 (SAS Institute Inc., Cary, NC, USA) and stata/mp 11.1 (StataCorp LP, College Station, TX 77845 USA).


Study population

A total of 1257 individuals accepted the invitation to take part in the study. Of those, 32 were not included because of previous CVD or heart valve disease, resulting in a study population of 1225 subjects. Because of inadequate scan quality (because of severe obesity) or lack of CT images (because of claustrophobia), the Agatston score was missing for four subjects (two men and two women). The prevalence of CAC (Agatston score >0) was 43% (523/1221 participants). The selection of the study population is illustrated in Fig. 1. CAC was less common in women compared to men (33% vs. 58%; < 0.0001) and in those aged 50 compared to 60 years (35% vs. 54%; < 0.0001). CAC was more pronounced in patients living in the city centre versus urban or rural areas, both in men (69% vs. 56%; = 0.009) and women (42% vs. 30%; = 0.02) as well as in 50-year-old (48% vs. 32%; = 0.001) and 60-year-old subjects (61% vs. 53%; = 0.09). Characteristics of the study cohort in relation to place of residence are shown in Table 1.

Figure 1.

 Participants in the DanRisk screening programme.

Air pollution

As shown for the city of Odense in Figs 2 and 3, the intensity of air pollution as estimated by NOx and PM10 concentrations differs significantly by location, with the highest values for the city centre. In Fig. 2, mean values for yearly NOx flux are shown for Odense city centre and the urban area; the NOx level for rural districts is also shown for comparison. Results from measurements of NOx can be followed back to 1994 with the same ratio between city centre, urban and rural districts until 2009 (Fig. 2).

Figure 2.

 Yearly measurements of mean of NOx according to place of residence [14].

Figure 3.

 Yearly measurements of mean PM10 according to place of residence [14].

Explanatory factors

Independent explanatory factors for the presence of CAC are shown in Table 2. Sex and age were most strongly associated with the presence of CAC with odds ratios (ORs) of 3.2 [95% confidence interval (CI) 2.5–4.2; < 0.0001] for men compared with women and 2.2 (95% CI 1.7–2.8; < 0.0001) for those aged 60 vs. 50 years. The OR for subjects living in the city centre versus those living outside the city centre was 1.8 (95% CI 1.3–2.4; = 0.0003).

Table 2.   Independent explanatory factors for the presence of coronary artery calcification
Risk factorOdds ratio95% Confidence intervalP value
  1. CVD, cardiovascular disease.

  2. a‘Air pollution’: traffic-intense areas chosen by postcode.

  3. bHypercholesterolaemia: use of lipid-lowering medicine or cholesterol level of ≥5 mmol L−1 or LDL cholesterol of ≥3 mmol L−1.

  4. cUse of antihypertensive medical treatment and/or blood pressure ≥140/90 mmHg at the time of examination.

Smoking status1.91.4–2.5<0.0001
Living in the city centrea1.81.3–2.40.0003
Family history of CVD1.51.1–2.00.01


We have shown that traditional risk factors for CVD and living in the city centre are independently associated with the presence of CAC in a cohort of asymptomatic middle-aged subjects. Men showed a three-fold increase in the risk of having CAC, compared with women, and the risk almost doubled for 60-year-old compared with 50-year-old subjects, participants with diabetes and smokers. Thus, living in the city centre almost doubles the risk of having CAC when compared to living outside the city centre.

CAC prevalence

The estimated prevalence of CAC in this study is comparable to reported data from two earlier studies of asymptomatic individuals. In the Multi-Ethnic Study of Atherosclerosis (MESA) cohort from the USA, the prevalence of CAC was slightly lower (41% and 16% of 50-year-old men and women and 68% and 36% of 60-year-old men and women, respectively). By contrast, the prevalence was slightly higher in the German Heinz Nixdorf Recall (HNR) study compared with the present study [15, 16]. Both these studies included the Framingham risk factors in an asymptomatic, nonreferred, gender-balanced cohort. The risk profiles between studies, however, are difficult to compare because the DanRisk cohort included only 50- and 60-year-old men and women, whereas the MESA cohort included individuals aged 45–84 years old and the HNR study included 45- to 75-year-old subjects. The Danish and German cohorts seem to have higher blood pressure and higher cholesterol levels, and more participants were current smokers, compared with the US cohort [15].

Explanatory factors

In the MESA study, hyperlipidaemia and hypercholesterolaemia were associated with an increased relative risk of 1.22 for the presence of CAC [17]. Moreover, conventional risk factors were associated with the risk of developing CAC amongst participants free of CAC at baseline [18]. In addition, LDL cholesterol was found to be an explanatory factor for incident cases but not in the model for progression of existing CAC. In a multivariate analysis in the HNR study, male gender, age (per 5-year increase), HDL cholesterol, LDL cholesterol, blood pressure and smoking remained independently associated with the presence of CAC [19].

Data on the presence of CVD risk factors and their association with CAC from the MESA and the HNR studies were evaluated by Erbel et al. [15]. A relative risk regression model was used to compare the presence of CAC in the two studies. The association between age and CAC was almost identical in the two cohorts with an approximately 47% increase in prevalent cases with each 10-year increase in age. A consistent finding in both the HNR and MESA cohorts as well as the present study is that age and sex were most strongly associated with the presence of CAC.

Living in the city centre

The level of human exposure to general air pollution can be associated with traffic and population density at specific locations; thus, a subject’s place of residence is often used as a surrogate for exposure to air pollution [10]. In the present study, even after adjustment for demographic and clinical variables, residency status was independently associated with the presence of CAC, and the prevalence of CAC was highest in people living in the city centre.

Noise, level of stress and educational or socio-economic factors may also influence CAC. However, in the present study, information on these variables was not available; thus, the possibility that they were differently distributed amongst the three groups (city centre, urban and rural dwellers) cannot be excluded. The most robust of these risk factors is noise. Noise [20, 21] as well as stress [22] has been shown to increase blood pressure levels. In the present study, however, mean blood pressure was lower for the group living in the city centre compared to those living outside. Accordingly, blood pressure was not significantly associated with the presence of CAC in this study. Furthermore, there were no differences in heart rate (another surrogate for stress) prior to scanning, with a mean (±SD) of 69 (±12) beats per min in those living in city centre and mean (±SD) of 69 (±12) living in the urban or rural, thus suggesting that sympathetic activity was equal in the two groups. However, the influence of various stress confounders on the association between residency and asymptomatic CVD needs delineation in future studies.

The mechanisms by which air pollution may contribute to the incidence of CAC are not well understood; it has previously been reported that traffic exposure or air pollution can trigger an acute cardiovascular event [23] but the results of the present study suggest the involvement of a different mechanism. In the HNR study, the relation between long-term traffic exposure and subclinical atherosclerosis was investigated [9, 10, 24], defined either by CAC or ankle–brachial index [24]. A validated chemical transport model (European air pollution dispersion) was used to assess exposure, and a specific value was assigned for the annual fine PM concentrations for each individual. The urban PM air pollution was related to both carotid intima–media thicknesses [9] and Agatston score [25] with stronger associations with particles <2.5 μm than with larger particles (PM10). Also, an association between air pollution and the risk of having an Agatston score above the age- and gender-specific 75th percentile was described [10]. By contrast, no consistent association between air pollution and CAC was found in the MESA study [11]. In the latter study, slightly different statistical methods were applied, measuring both the relative prevalence of CAC (binomial regression) and the amount of CAC (linear regression). The estimate of air pollution was very similar in the MESA and HNR studies. The lower overall number of participants with Agatston score >0 in the MESA study, compared with the HNR study, might be a reasonable explanation for the differences in association between air pollution and the presence of CAC. However, the differences between the studies may also be explained by variations in living conditions in the USA and Europe. Our results seem to be in line with the German observations, although our focus was on prevalent cases, whereas the exact value of Agatston score was considered in the HNR study. The estimate of traffic pollution used in the HNR and MESA studies may be a more accurate estimate of air pollution. However, the surrogate of air pollution used in this study is commonly applied and easily accessible. Thus, it should be acknowledged that results on air pollution from different studies are not directly comparable. The fact that people living in the city centre are more exposed to pollution as compared to those living outside the city centres is supported by other reports including measurements from the Danish National Environmental Research Institute, showing almost triple the concentration of NOx in city centres as compared to urban districts [14].

There are some limitations that should be discussed. First, the Agatston score is a precise estimate of the atherosclerotic burden, but CAC only represents part of the atherosclerotic lesion. It should, however, be acknowledged that studies have shown that the presence of coronary plaque may occur in patients with an Agatston score of zero [26]. Nevertheless, an Agatston score of zero is generally associated with an excellent prognosis [26].

Second, atherosclerosis develops over many years during which some participants may have moved between polluted and nonpolluted areas, leading to misclassification. A detailed recall study of residency over time is needed to examine the existence of a dose–response relation between ‘air pollution’ and the Agatston score. There were too few participants in the present study for subanalyses according to various Agatston score intervals. The retrospective exposure assessment defined by taking the lifetime place(s) of residence of the participants into account inevitably leads to some degree of misclassification, and the lack of residential history is a limitation of our study.

Third, there was an association between residential status and the presence of CAC even after adjustment for traditional CVD risk factors. However, other factors potentially related to place of residence, such as social and educational status, lifestyle and exposure to noise, were not taken into account. Noise is known to cause a disposition towards higher blood pressure [20]; however, we found a lower average blood pressure amongst people living in city centres. Moreover, continuous surveillance of air pollution is carried out in Denmark and shows that, in major cities that are comparable to the cities included in this study, pollution is related to place of residence with much higher mean values of pollutants in the long term in city centres.

In conclusion, the presence of CAC in asymptomatic ‘healthy’ middle-aged subjects is related to the presence of traditional CVD risk factors and living in a city centre (a surrogate for air pollution). These findings may be relevant from a public health perspective, but should be verified in further studies.

Conflict of interest statement

None of the authors has any conflicts of interest to declare.


We are indebted to the Department of Cardiology and the Department of Nuclear Medicine, Odense University Hospital, the Department of Cardiology, Vejle Hospital and the Department of Cardiology, Esbjerg Hospital.