OUP user menu

Relationship between chronic obstructive pulmonary disease and subclinical coronary artery disease in long-term smokers

Thomas Rasmussen, Lars Køber, Jesper Holst Pedersen, Asger Dirksen, Laura Hohwü Thomsen, Steen Stender, John Brodersen, Jaap Groen, Haseem Ashraf, Klaus Fuglsang Kofoed
DOI: http://dx.doi.org/10.1093/ehjci/jet057 1159-1166 First published online: 2 May 2013

Abstract

Aims Cardiovascular conditions are reported to be the most frequent cause of death in patients with chronic obstructive pulmonary disease (COPD). However, it remains unsettled whether severity of COPD per se is associated with coronary artery disease (CAD) independent of traditional cardiovascular risk factors. The aim of this study was to examine the relationship between the presence and severity of COPD and the amount of coronary artery calcium deposit, an indicator of CAD and cardiac risk, in a large population of current and former long-term smokers.

Methods and results In this cross-sectional study, long-term smokers without clinically manifested CAD were recruited from the Danish Lung Cancer Screening Trial and classified according to lung function by the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria. Coronary artery calcium deposit as a measure of subclinical CAD and cardiac risk was evaluated with multi detector computed tomography and the Agatston coronary artery calcium score (CACS). Participants were categorized into five CACS risk classification groups according to the CACS. The population (n = 1535) consisted of 41% participants without COPD, 28% with mild, and 31% with moderate-to-severe COPD (n = 46 with severe COPD). In addition to age, male gender, hypertension, hypercholesterolaemia, and continued smoking, COPD according to GOLD classification were independent predictors of a higher CACS risk classification group in multivariable analysis [odds ratio (OR): 1.28 (1.01–1.63) and OR: 1.32 (1.05–1.67), for mild and moderate-to-severe COPD, respectively, compared with no COPD].

Conclusion COPD in long-term smokers is independently correlated with the CACS, while COPD severity per se does not show a dose–response relationship.

  • Coronary artery calcium score
  • Chronic obstructive pulmonary disease
  • Smoking
  • Risk assessment

Introduction

Chronic obstructive pulmonary disease (COPD) is mainly observed as a consequence of long-term smoking and is the fourth leading cause of death worldwide.1,2 Although patients with COPD may die from respiratory failure and lung cancer, coronary artery disease (CAD) plays a significant role as cardiovascular conditions have been reported to be the most frequent cause of death in COPD.35 Earlier studies have investigated the relationship between COPD and cardiovascular mortality.6,7 However, to what extent COPD per se, independent of traditional cardiovascular risk factors, is associated with an increased risk of cardiovascular morbidity and mortality remains unsettled. The most advocated prognostic models in COPD—the Body mass index, airflow Obstruction, Dyspnoea, and Exercise index8 and the simplified Age, Dyspnoea and airflow Obstruction index9 do not include markers of cardiovascular risk. Similarly, acknowledged models for cardiovascular risk assessment in subjects without symptoms of ischaemic heart disease, such as the Framingham Risk Score,10 do not include COPD as a predictor of risk.

The risk of future cardiac events in subjects without cardiac symptoms can be estimated by the recording of calcifications in the coronary arteries using multi detector computed tomography (MDCT) and quantified as a coronary artery calcium score (CACS). The detection of calcifications in the coronary arteries documents the presence of subclinical CAD. The CACS measured by the Agatston scoring method11 has been extensively examined and documented as a strong predictor of future cardiac events apparently independent of Framingham risk factors in persons with no symptoms of heart disease.1217

The aim of this study was to assess the relationship between the presence and the severity of COPD and subclinical CAD assessed by the CACS in long-term smokers.

Methods

Study population and design

Participants for our study were recruited from The Danish Lung Cancer Screening Trial (DLCST; www.ClinicalTrials.org, registration number: NCT00496977), a randomized-controlled population screening trial, in which 4104 men and women were randomized to either MDCT scan and spirometry (MDCT-screening group) or no screening and spirometry (control group).18 The study was conducted in a single institution. Participants volunteered in response to advertisements in local media that provided information about the study. Inclusion criteria for the trial were age between 50 and 70 years, current or former smoking with at least 20 pack-years, forced expiratory volume in first second (FEV1) >30% of predicted. Exclusion criteria were body weight >130 kg or previous treatment for any kind of cancer within 5 years, tuberculosis within 2 years or any serious illness with life expectancy <10 years. The study was approved by the National Ethics Committee of Denmark (identification no. H-KA-02045, supplementary protocol 20148) and all participants gave written informed consent.

The present cross-sectional DLCST substudy includes participants from the MDCT-screening group who underwent spirometry and MDCT at the screening round the year 2009–2010, n = 1864. A detailed cardiovascular history, clinical examination, and laboratory tests with focus on cardiac risk [cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), very low-density lipoprotein (VLDL), and triglycerides] were recorded for every participant. Participants identified with clinically manifested CAD defined as previous coronary artery bypass graft, percutaneous coronary intervention, history of acute myocardial infarctions (n = 88), or angina pectoris by Rose Angina Questionnaire19 (n = 47) were excluded from this substudy. Furthermore, participants with neither normal nor obstructive lung function (defined below, n = 135), persons with implanted cardiac pacemaker resulting in image artefacts (eight participants), poor MDCT image quality or missing MDCT image slices (n = 35), and participants with missing spirometry data (n = 16) were excluded from the substudy.

Spirometry

Spirometry was performed using a computerized system (Spirotrac IV; Vitalograph, Buckingham, UK), and expressed in absolute values and as a percentage of predicted values according to European reference equations.20 The FEV1 and the forced vital capacity (FVC) measured in litres were determined as the best of three measures with <5% variation. A FEV1/FVC ratio of ≥70% and a FEV1 in per cent of predicted (FEV1%) of >80% were considered as normal lung function, whereas a FEV1/FVC ratio of <70% was classified as COPD. Lung function was considered neither normal nor obstructive, if the FEV1/FVC ratio was >70% and FEV1% was <80%. The severity of COPD was classified according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria21 by FEV1 in per cent of predicted (FEV1%). Participants were categorized as having mild COPD, when FEV1% was ≥80% and moderate-to-severe COPD when FEV1% was >30% and <80%.22

MDCT imaging and image analysis

All CT scans were performed with a 16-row Philips Mx 8000 MDCT scanner, Philips Medical Systems, Eindhoven, the Netherlands. Scans were performed supine after full inspiration with caudocranial scan direction including the entire ribcage and upper abdomen with a low-dose technique, 120 kV and 40 mAs. Scans were performed with spiral data acquisition with the following parameters: section collimation 16 × 0.75 mm, pitch 1.5, rotation time 0.5 s. The obtained data were reconstructed with a section width of 3 mm and a soft kernel algorithm. All image data were stored in DICOM format.

The CACS was assessed by one observer using Vitrea v. 6.0 (Vital Images, Inc., MN, USA). A standardized procedure for calcium scoring with clear-cut definitions of the delineation of the coronary arteries was designed to lower the variability in image analysis. According to the Agatston scoring method, an attenuation threshold of 130 Hounsfield units (HU) was used to identify coronary calcifications. The sensitivity of this approach was evaluated in a dedicated phantom containing 100 calcifications with different diameters (between 0.5 and 2.0 mm) and densities (between 90 and 540 HU). The phantom was scanned using the aforementioned protocol and the measured Agatston score was compared with reference values derived from phantom scans obtained using electron beam computed tomography (EBCT), the GOLD standard in calcium scoring. The phantom EBCT scans were performed with an Imatron C300, GE, Milwaukee, WI, USA using the following parameters: 130 kV, 600 mAs, collimation 3 mm, data reconstruction with a sharp kernel with a slice thickness and an increment of 3 mm. Using our MDCT imaging approach, calcifications with diameters ≥1.2 mm and hydroxyappatite densities ≥240 mg/ml were accurately detected (attenuation threshold 130 HU), compared with similar measures performed by EBCT (data not shown). Agatston scores derived in our participants were evaluated by intra- and inter-observer analysis of a subset of 100 random scans in the Agatston score range from 1 to 1913. Intra- and inter-observer Agatston score risk categorization (defined below) correlated well (kappa statistic: κ = 0.99 and κ = 0.97, respectively).

Participants were categorized into five CACS cardiac risk classification groups according to the CACS (CACS = 0, CACS = 1–10, CACS = 11–100, CACS = 101–400, and CACS > 400) as suggested by Rumberger,23 and based upon recommendations from the American Heart Association's Prevention Conference V.24 The 10-year risk of incident CAD events in asymptomatic subjects ranges from 1 to 2% in subjects with CACS = 0 to 20–30% in subjects with CACS > 400.25

Cardiovascular risk factors and biomarkers

Information on cardiovascular risk factors was collected, including smoking status, family history of premature coronary heart disease, medical treatment for hypertension, hypercholesterolaemia, and diabetes. The level of LDL, HDL, VLDL, cholesterol, triglycerides and plasma glucose in addition to high-sensitivity CRP (hs-CRP), creatinine, and cystatin C was measured.

The current smoking status was classified according to self-report. The pack-years were calculated as the product of the total years of smoking and the average cigarette consumption per day divided by 20. Participants were defined as having diabetes if they were medically treated with oral hypoglycaemic agents or subcutaneous insulin or if a non-fasting blood glucose value exceeded 11 mmol/L (198 mg/dL). Participants were considered having hypertension if they were medically treated with antihypertensive agents. Participants were considered having hypercholesterolaemia if they were medically treated with statins or if their non-fasting total cholesterol was ≥6.2 mmol/L (240 mg/dL), LDL ≥4.1 mmol/L (160 mg/dL), triglycerides ≥2.3 mmol/L (200 mg/dL), or HDL <1.0 mmol/L (40 mg/dL) based upon recommendations from the National Cholesterol Education Program.26 A family history of premature CAD was considered present if the participant had a first-degree relative who had a major cardiovascular event (myocardial infarction, percutaneous coronary intervention, coronary artery bypass graft, or sudden cardiac death) before the age of 55 years in men and 65 years in women. The body mass index was calculated as the weight in kilograms divided by the square of the height in metres.

Statistical analysis

Categorical variables were expressed as percentages and continuous variables were reported as medians and inter-quartile ranges. Differences in baseline characteristics between CACS risk classification groups were assessed with the Cochran–Armitage trend test for discrete variables and the Kruskal–Wallis test for continuous variables. The χ2 test was used to assess differences between COPD subgroups (no COPD, mild COPD, and moderate-to-severe COPD) in CACS risk groups. Log2(CACS + 1) was used (because of the skewed CACS data) to illustrate the relationship to FEV1%. A logistic regression model was used for multivariable analysis to determine the associations between CACS risk groups and clinical variables. Classical risk factors for CAD were included in the model. The model was, furthermore, adjusted for age, gender, hypertension, hypercholesterolaemia, diabetes, smoking status, and COPD severity. A two-tailed P-value <0.05 was considered statistically significant. Statistical analyses were performed using SAS® for Windows, version 9.1 (SAS institute, Cary, NC, USA).

Results

Characteristics of participants

The study population consisted of 1535 participants (median age 61 years, 45% women; Table 1). Participants had a median of 33 pack-years and 60% were current smokers. COPD was found in 59% of all participants with 28% having mild COPD, and 31% having moderate-to-severe COPD (severe COPD accounted for 10% in participants with moderate-to-severe COPD) according to GOLD classification.

View this table:
Table 1

Characteristics of participants

Total (n = 1535)
Age (years)61 (57–65)
Female gender (%)45
Clinical history of
 Hypertension (%)23
 Stroke (%)3
 Periferal artery disease (%)9
 Diabetes (%)a4
 Hypercholesterolaemia (%)b57
Family history of premature CAD (%)13
Current smokers (%)60
Pack-years (years)33 (27–42)
BMI (kg/m2)25 (23–28)
Waistline (cm)94 (86–101)
Systolic blood pressure (mmHg)143 (130–158)
Heart rate (/min)71 (64–79)
Lung function
 FEV1 (l)2.51 (2.06–3.02)
 FEV1% (%)87 (76–96)
 FVC (l)3.69 (3.08–4.37)
 FEV1/FVC ratio (%)68 (63–74)
 No COPD (%)41
 Mild COPD (%)28
 Moderate-to-severe COPD (%)31
Biomarkers
 Creatinine (µmol/L)74 (64–83)
 Cystatin-C (mg/L)0.93 (0.85–1.02)
 Glucose (mmol/L)5.2 (4.8–5.7)
 Cholesterol (mmol/L)5.7 (5.0–6.3)
 LDL (mmol/L)3.3 (2.6–3.9)
 HDL (mmol/L)1.54 (1.27–1.83)
 VLDL (mmol/L)0.7 (0.5–1.0)
 Triglycerides (mmol/L)1.53 (1.07–2.22)
 hs-CRP (mg/L)1.7 (0.8–3.5)
  • Continuous variables are expressed as median and inter-quartiles in parenthesis. Categorical variables are expressed as percentages.

  • CAD, coronary artery disease; BMI, body mass index; FEV1, forced expiratory volume in first second; FEV1%, forced expiratory volume in first second in per cent of predicted; FVC, forced vital capacity; COPD, chronic obstructive pulmonary disease; LDL, low-density lipoprotein; HDL, high-density lipoprotein; VLDL, very low-density lipoprotein; hs-CRP, highly sensitive C-reactive protein.

  • aAntidiabetic treatment and/or non-fasting blood glucose >11 mmol/L (198 mg/dL).

  • bLipid-lowering therapy and/or total cholesterol ≥6.2 mmol/L (240 mg/dL), LDL ≥ 4.1 (160 mg/dL), triglycerides ≥2.3 (200 mg/dL), or HDL < 1.0 (40 mg/dL).

CACS risk classification groups

Distribution of participants in CACS risk classification groups is given in Table 2. Participants with a higher CACS were more likely to be older and of male gender and were more likely to have a clinical history of stroke, peripheral artery disease, diabetes, hypertension, or hypercholesterolaemia (P < 0.01). Similarly, a higher CACS was more likely to be found in current smokers and participants with more pack-years (P < 0.01). No difference in glucose, creatinin, cystatin-c, HDL, VLDL, triglycerides, and hs-CRP between CACS risk classification groups was noted, whereas cholesterol and LDL tended to be lower in participants with a higher CACS due to treatment with statins (data not shown).

View this table:
Table 2

CACS risk classification groups in relation to patient characteristics

 CACSP-value
0 (n = 651)1–10 (n = 157)11–100 (n = 322)101–400 (n = 244)>400 (n = 161)
Age (years)59 (56–63)61 (58–64)62 (58–65)63 (59–66)65 (62–68)<0.01
Female gender (%)6050382723<0.01
Clinical history of
 Hypertension (%)1719312339<0.01
 Stroke (%)113312<0.01
 Peripheral artery disease (%)81071118<0.01
 Diabetes (%)a235311<0.01
 Hypercholesterolaemia (%)b4966626068<0.01
Family history of premature CAD (%)12181310170.57
Current smokers (%)5563606667<0.01
Pack-years (years)31 (25–38)33 (27–43)35 (28–43)36 (29–45)41 (31–50)<0.01
BMI (kg/m2)25 (23–28)25 (23–28)26 (23–29)26 (23–28)24 (22–28)<0.01
Waistline (cm)91 (83–99)93 (86–101)95 (88–103)95 (89–102)94 (88–102)<0.01
Systolic blood pressure (mmHg)140 (126–153)145 (133–163)145 (134–157)147 (133–162)144 (132–160)<0.01
Heart rate (/min)70 (64–78)73 (66–80)72 (64–80)72 (64–81)71 (63–81)0.33
  • Continuous variables are expressed as median and inter-quartiles in parenthesis. Categorical variables are expressed as percentages.

  • CAD, coronary artery disease; BMI, body mass index.

  • aAntidiabetic treatment and/or non-fasting blood glucose >11 mmol/L (198 mg/dL).

  • bLipid-lowering therapy and/or total cholesterol ≥6.2 mmol/L (240 mg/dL), LDL ≥4.1 (160 mg/dL), triglycerides ≥2.3 (200 mg/dL), or HDL <1.0 (40 mg/dL).

CACS in relation to lung function

The relationship between CACS and FEV1% in men and women is shown in Figure 1 (the figure was divided into men and women because of large CACS differences between genders). The CACS was inversely related to FEV1% (P < 0.01).

Figure 1

(A) Relationship between CACS and FEV1% in women (P < 0.01). Please note that the median for FEV1% >75 is 0. (B) Relationship between CACS and FEV1% in men (P < 0.01).

The distribution of COPD according to GOLD classification in CACS risk groups is shown in Figure 2. A significant linear trend was found between COPD severity and CACS risk classification (P < 0.01).

Figure 2

Distribution of lung function according to CACS groups (P < 0.01).

Predictors of CACS risk classification groups

Predictors of CACS risk classification assessed by univariable and multivariable analysis are listed in Table 3. By multivariable analysis age, male gender, smoking status, hypercholesterolaemia, hypertension, and COPD according to GOLD classification were found to be significant independent predictors of CACS risk categorization. COPD severity per se did not show a dose–response relationship.

View this table:
Table 3

Predictors of CACS cardiovascular risk groups

VariableUnivariable analysisMultivariable analysis
OR (95% CI)P-valueOR (95% CI)P-value
Age1.13 (1.11–1.16)<0.011.12 (1.10–1.14)<0.01
Male gender3.03 (2.50–3.67)<0.012.94 (2.41–3.60)<0.01
Diabetes2.48 (1.56–3.95)<0.011.34 (0.82–2.17)0.24
Hypertension1.84 (1.48–2.29)<0.011.44 (1.14–1.82)<0.01
Hypercholesterolaemia1.60 (1.32–1.94)<0.011.51 (1.23–1.84)<0.01
Current smoker1.38 (1.14–1.66)<0.011.49 (1.22–1.82)<0.01
Mild COPDa1.32 (1.06–1.66)0.011.28 (1.01–1.63)0.04
Moderate-to-severe COPDa1.66 (1.34–2.07)<0.011.32 (1.05–1.67)0.02
  • The ORs of being in a higher CACS cardiovascular risk group are given. OR, odds ratio; 95% CI, 95% confidence interval.

  • aWith no COPD as a reference.

Discussion

The results of this study demonstrate that the presence of COPD in long-term smokers is independently associated with the presence of subclinical CAD measured by coronary artery calcium deposit. The CACS was significantly higher in subgroups with COPD compared with subgroups without COPD. Furthermore, the CACS was significantly higher in more severe COPD subgroups compared with less severe. Finally, COPD according to GOLD classification was found to be a predictor of CACS categorization independent of known cardiovascular risk factors including the smoking status. Although, COPD severity per se does not show a dose–response relationship, this implies that COPD might be considered a risk factor of CAD.

Although the mechanisms responsible for the association between COPD and CAD are not fully understood, some pathophysiological determinants such as, especially, smoking are shared.27,28 A common pathogenetic feature of COPD and CAD is the presence of a low-grade systemic inflammatory process, which has been hypothesized to cause the loss of elastin in both the arterial wall and the alveoli causing arterial wall stiffness and loss of alveolar tissue.27,2931 Another pathogenetic feature suggested is that the endothelial dysfunction responsible for insults in the coronary arteries also affects the pulmonary vasculature.32,33 This can, in part, serve as explanation for the association between the atherosclerosis and impaired lung function in COPD. These hypotheses suggest a process where impairment of lung function and development of atherosclerosis evolve in parallel in the presence of systemic inflammation.

The results of the present study corroborate results obtained in previous studies. Newman et al.34 studied 614 men and women (≥65 years of age), as part of the National Health Study. Besides ascertainment of cardiovascular risk factors, they performed EBCT with the purpose to assess the CACS. They found a prevalence of self-report COPD in the lowest CACS quartile of 11% in men and 22% in women, while the prevalence in the highest CACS quartile was 22 and 39%, respectively. In our study, the prevalence of COPD according to GOLD classification was 54% in participants with CACS = 0 compared with 70% in participants with CACS > 400. The relative higher fraction of participants with COPD in our study compared with Newman et al. is related to the nature of study participants. The participants in the present study were all current or former long-term smokers with at least 20 pack-years, whereas participants in the study from Newman et al. were all a mixture of people who had never smoked, former smokers, and light and heavy smokers. Furthermore, Newman et al. relied on self-report information, whereas the present study relied on spirometry data.

The present study suggests COPD as an independent predictor of the CACS, which corroborate with previous studies. In the Lung Health Study7, 5.887 smokers (35–60 years of age) with mild-to-moderate airway obstruction were randomized to either usual care plus placebo, special intervention for smoking cessation plus placebo, or special intervention for smoking cessation plus ipratropium. During the 5-year follow-up, 2.5% of the original cohort died, 25% of these from a cardiovascular event. Approximately 13% of the cohort had at least one hospitalization during the 5-year follow-up of which cardiovascular events accounted for 42%. For every 10% decrease in FEV1, all-cause mortality increased by 14%, cardiovascular mortality increased by 28%, and non-fatal coronary events increased by almost 20%, after adjustment for age, gender, smoking status, and treatment assignment. In another study of 1.195 men and women (20–89 years of age) with a follow-up of 29 years, Schünemann et al.35 reported a relative risk (RR) of cardiovascular mortality of 1.96 (95% CI: 0.99–3.88) in women and 2.11 (95% CI: 1.20–3.71) in men when comparing the lowest FEV1 quintile with the highest quintile. Hole et al.36 found in a study of 15,411 men and women (45–64 years of age at baseline) with a follow-up of 15 years, a RR of mortality related to cardiovascular disease of 1.88 (95% CI: 1.44–2.47) in women and 1.56 (95% CI: 1.26–1.92) in men, comparing the lowest quintile of FEV1 with the highest quintile. When comparing the lowest quintile of FEV1 with the highest, the population attributable risk for ischaemic heart disease-related deaths because of reduced FEV1 was 24% (95% CI: 14–34%) in women and 26% (95% CI: 19–34%) in men, independent of the burden related to smoking. The magnitude of the mortality burden because of reduced FEV1 was similar to the burden imposed by hypercholesterolaemia.

The findings in the present study, that the extent of atherosclerosis relates to impaired lung function, is supported by Zureik et al.37 who studied the association between the peak expiratory flow and the occurrence of atherosclerotic plaques in the extracranial carotid arteries over a 4-year period in 656 subjects (59–71 years of age) without CAD and stroke at baseline. The unadjusted OR of a carotid atherosclerotic plaque in the lowest quintile of the relative peak expiratory flow was 3.07 (95% CI, 1.62–5.85) (P < 0.001 for trend) compared with the highest quintile and remained highly significant after adjustment for known major cardiovascular risk factors. In another study, Zureik et al.38 studied 139 men (30–70 years of age) free of CAD. A correlation between impaired lung function and increased carotid-femoral pulse-wave velocity, a measure that increases proportionally with an increase in aortic stiffness, was demonstrated. For every 1 SD increase in pulse-wave velocity (2.5 m/s), FEV1 decreased by 195.2 ± 50.1 mL (P < 0.001) in an age- and height-adjusted analysis. Further adjustment for cardiovascular risk factors did not markedly alter the results. These findings corroborate with a recent substudy of the National Lung Screening Trial (NLST), reporting a relationship between thoracic aortic calcification and emphysema.28

Although the results of the present study are in concordance with earlier published studies, they are in contrast to a few. In a recent cross-sectional case–control study from Izquierdo et al.,39 with 204 patients hospitalized with CAD (stable angina pectoris, unstable angina pectoris, or acute myocardial infarction) in the stable phase and 100 control hospital patients, COPD was not associated with CAD, OR after adjustment for cardiovascular risk factors was 1.14 (95% CI: 0.57–2.29). Although patients with CAD had significantly poorer lung function than the controls, the relationship could solely be explained by a higher cardiovascular risk factor profile in patients with COPD. The finding might be related to a potentially underpowered study. In a recent substudy of 3,642 participants in the MESA study, Barr et al.40 examined the association between lung function and measures of subclinical atherosclerosis by carotid-intima media thickness, ankle-branchial index, and CACS. Decrements in FEV1 and FEV1/FVC ratio was associated with increased carotid-intima media thickness in smokers (P = 0.03 and P < 0.001, respectively), whereas the reduced ankle-branchial index was associated to per cent emphysema regardless of smoking history (P = 0.004). Furthermore, which is in conflict with this present study, they concluded that neither emphysema nor COPD (prevalence ratio for severe airway obstruction 0.99 (95% CI: 0.91–1.07) was associated with the CACS. An explanation to this conflict could rely on the inclusion criteria. In the present study, we only included long-term smokers, whereas the MESA study was a population-based study including both never smokers (47%) and former (39%) and current smokers (13%) of different extent.

Our study shows that COPD in smokers is associated with increased calcium deposits in the coronary arteries independent of known cardiac risk factors. This suggests that COPD might be considered a predictor in cardiovascular risk assessment models in smokers.

Limitations

The inclusion of participants in response to advertisements in the written media pose a potential bias for selection that favour the more enlightened and possibly less sick smokers. It is of importance to note that the findings of the present study are limited to a population of heavy smokers and thus smoking induced COPD.

It is a potential limitation that CT scans were performed without ECG-gating. Nevertheless, comparisons of the CACS obtained by gated vs. ungated MDCT shows a high degree of concordance, which makes categorization in Agatston CACS groups and thus the assessment of the risk of future adverse cardiac events for subjects in lung cancer screening cohorts possible.41,42 On the basis of our phantom measurements, we know that our 16-slice MDCT scanner used for this study underestimates smaller calcifications compared with EBCT. Both methods fail to detect calcifications with diameters smaller than ∼1.2 mm, but the limit of detection is lower for EBCT. Consequently, our method may miss smaller calcifications with densities in the range of 130–240 mg/ml compared with EBCT. Therefore, a fraction of subjects categorized with CACS = 0 in our study could have had detectable calcifications if scanned with EBCT. Nevertheless, this finding seems unimportant for the higher CACS categories where this very small deviation will not result in a change of CACS risk classification.

In this substudy, we did not measure cardiovascular outcomes but rather CAD severity by the CACS, a direct measure of CAD.

Data on the socioeconomic status and the physical activity level were not included in this substudy. Furthermore, the results of the present study are limited to reflect long-term smokers.

Conclusion

COPD in long-term smokers is independently correlated with the CACS, while COPD severity per se does not show a dose–response relationship.

Funding

T.R. was supported by an unrestricted grant from AstraZeneca AB. The DLCST trial was funded in full by a governmental grant by the Danish Ministry of Health and Prevention from 2004 to 2011.

Conflict of interest: none declared.

Acknowledgements

The authors thank the investigators, staff, and participants of the Danish Lung Cancer Screening Trial for their valuable contributions.

References

View Abstract