Inflammatory Biomarkers and Comorbidities in Chronic Obstructive Pulmonary Disease
Mette Thomsen1,2, Morten Dahl1,2,3, Peter Lange2,4,5,6, Jørgen Vestbo7,8, and Børge G. Nordestgaard1,2,4
1Department of Clinical Biochemistry and the Copenhagen General Population Study, Herlev Hospital, Copenhagen University Hospital, Herlev, Denmark; 2 Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark; 3 Department of Clinical Biochemistry, Rigshospitalet Copenhagen University Hospital, Copenhagen, Denmark; 4 The Copenhagen City Heart Study, Bispebjerg Hospital, Copenhagen University Hospital, Copenhagen, Denmark; 5 Respiratory Section, Hvidovre Hospital, Copenhagen University Hospital, Hvidovre, Denmark; 6 Department of Public Health, Section of Social Medicine, University of Copenhagen, Copenhagen, Denmark; 7 Department of Respiratory Medicine, Odense University Hospital, University of Southern Denmark, Odense, Denmark; and 8 Respiratory Research Group, Manchester Academic Health Science Centre, University Hospital South Manchester National Health Service Foundation Trust, Manchester, United Kingdom
Rationale: Patients with chronic obstructive pulmonary disease (COPD) have evidence of systemic inflammation that may be implicated in the development of comorbidities.
Objectives: To test the hypothesis that elevated levels of three inflammatory biomarkers are associated with increased risk of comorbidities in COPD.
Methods: We examined 8,656 patients with COPD from two large Danish population studies and during amedian 5 years’ follow-up recorded hospital admissions due to major comorbidities as endpoints.
Measurements and Main Results: We measured baseline C-reactive protein (CRP), fibrinogen, and leukocyte count, and recorded admissions due to ischemic heart disease, myocardial infarction, heart failure, type II diabetes, lung cancer, pneumonia, pulmonary embolism, hip fracture, and depression for all participants. Multifactorially adjusted risk of ischemic heart disease was increased by a factor of 2.19 (95% confidence interval, 1.48–3.23) in individuals with three biomarkers elevated (CRP . 3 mg/L, fibrinogen . 14 mmol/L, and leukocyte count . 9 3 10 9 /L) versus individuals with all three biomarkers at or below these limits. Corresponding hazard ratios were 2.32 (1.34– 4.04) for myocardial infarction, 2.63 (1.71–4.04) for heart failure, 3.54 (2.03–6.19) for diabetes, 4.00 (2.12–7.54) for lung cancer, and 2.71 (2.03–3.63) for pneumonia. Therewere no consistent differencesin risk of pulmonary embolism, hip fracture, or depression as a function of these three biomarkers.
Conclusions: Simultaneously elevated levels of CRP, fibrinogen, and leukocytecount are associatedwith a two- to fourfoldincreased riskofmajor comorbidities in COPD. These biomarkers may be an additional tool for clinicians to conduct stratified management of comorbidities in COPD.
Keywords: chronic obstructive pulmonary disease; comorbidities; in- flammation; biomarkers
(Received in original form June 25, 2012; accepted in final form August 23, 2012)
Supported by Herlev Hospital, Copenhagen University Hospital, The Danish Lung Foundation, The Danish Heart Foundation, Copenhagen County Foundation, and University of Copenhagen, all from Denmark. The funding sources had no role in the design and conduct of the study, in the collection, management, analysis, and interpretation of the data, or in the preparation, review, or approval of the manuscript.
Author Contributions: M.D., P.L., J.V., and B.G.N. designed the study. M.T. collected data and performed all analysis. M.D., P.L., J.V., and B.G.N. oversaw all analysis and contributed to the interpretation of data. M.T. wrote the first draft of the paper. M.D., P.L., J.V., and B.G.N. edited the paper, and all authors approved this paper in its final form.
Correspondence and requests for reprints should be addressed to Børge G. Nordestgaard, M.D., D.M.Sc., Department of Clinical Biochemistry, Herlev Hospital, Copenhagen University Hospital, Herlev Ringvej 75, DK-2730 Herlev, Denmark. E-mail: Boerge.Nordestgaard@regionh.dk
This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org
Am J Respir Crit Care Med Vol 186, Iss. 10, pp 982–988, Nov 15, 2012 Copyright ª 2012 by the American Thoracic Society Originally Published in Press as DOI: 10.1164/rccm.201206-1113OC on September 13, 2012 Internet address: www.atsjournals.org
Chronic obstructive pulmonary disease (COPD) is associated with an enhanced inflammatory response by the lungs to inhaled particles and gases, particularly cigarette smoke (1). Also, patients with COPD have increased levels of circulating cytokines, acute phase proteins, and inflammatory cells, indicating the presence of additional systemic inflammation (2–7). Such systemic inflammation may be implicated in the development of comorbidities in COPD (5, 6), such as cardiovascular disease, diabetes, lung cancer, pneumonia, pulmonary embolism, hip fracture, and depression (8–12). C-reactive protein (CRP), fibrinogen, and leukocyte count are biomarkers of systemic inflammation that are commonly used to monitor disease in patients with COPD.
Also, these biomarkers have in a number of studies been reported to be elevated in COPD (3). Thus, elevated levels of any or all of these three biomarkers may be associated with increased risk of development of comorbidities in patients with COPD. We tested the hypothesis that elevated levels of CRP, fibrinogen, and/or leukocyte count associate with increased risk of comorbidities in COPD. For this purpose we measured baseline CRP, fibrinogen, and leukocyte count in 8,656 patients with COPD from two large population-based studies consisting of more than 70,000 participants. During a median of 5 years’ follow-up, we recorded hospital admissions and deaths due to ischemic heart disease, myocardial infarction, heart failure, diabetes, lung cancer, pneumonia, pulmonary embolism, hip fracture, and depression for use as endpoints. We examined combinations of the three biomarkers and stratified analyses according to grades of COPD defined in accordance with Global Initiative for Chronic Obstructive Lung Disease (GOLD) (1). Finally, we calculated absolute 5-year risks and in addition retested the hypothesis using the lower limit of normal in the definition of COPD as a sensitivity analysis.
Methods
Participants
We studied age-stratified randomly selected white individuals from the Copenhagen City Heart Study and the Copenhagen General Population Study (13–15), two similar studies recruiting individuals from the adult Danish general population; there was no overlap of participants between studies. Individuals were selected on the basis of the national Danish Civil Registration System to reflect the population aged 20 to 100 years. The studies were approved by Herlev Hospital and Danish ethical committees. Written informed consent was obtained from all participants. The participants were analyzed as one collective cohort to obtain maximal statistical power.
The Copenhagen City Heart Study is a prospective population study initiated in 1976. At each examination, participants filled out a questionnaire and had physical measurements taken. During the 1991 to 1994 and the 2001 to 2003 examination, 9,995 individuals had spirometry performed and plasma levels of CRP and fibrinogen measured. In addition, 1,753 of the individuals from the 2001 to 2003 examination had leukocyte count measured.
The Copenhagen General Population Study is a prospective population study initiated in 2003 and still recruiting participants. Data on each participant collected in this study are identical to those in the Copenhagen City Heart Study. Between 2003 and 2009, 60,005 individuals had spirometry performed and plasma levels of CRP, fibrinogen, and leukocyte count measured.
After excluding 475 individuals younger than 40 years with selfreported asthma, we had 69,525 individuals available for analyses in the two studies. Of these, 8,656 individuals had COPD, defined as a ratio between FEV1 and FVC less than 0.7, and had all three biomarkers measured. Numbers of individuals included in analysis of each biomarker separately and in combination are seen in Table E1 in the online supplement.
Spirometry
FEV1 and FVC were determined without inhalation of a bronchodilator using a dry wedge spirometer (Vitalograph; Maids Moreton, Buckinghamshire, UK) in the Copenhagen City Heart Study and an EasyOne Spirometer (ndd Medizintechnik, Zurich, Switzerland) in the Copenhagen General Population Study. Reference values for FEV1 and the lower limit of normal for FEV1/FVC (5th percentile of a frequency distribution) were internally derived for men and women separately in a subsample of healthy never smokers using linear and quantile regression with age and height as covariates. COPD was defined as FEV1/FVC less than 0.7 and was grouped according to GOLD (1): GOLD 1 to 2 with FEV1% predicted greater than or equal to 50% and GOLD 3 to 4 with FEV1 % predicted less than 50%. For sensitivity analyses, COPD was defined as FEV1/FVC below the lower limit of normal.
Inflammatory Biomarkers
Plasma levels of high-sensitivity CRP, fibrinogen, and whole blood leukocyte count were measured using standard hospital assays at a central laboratory. The samples were analyzed in real time except for plasma levels of CRP in the 1991 to 1994 examination of the Copenhagen City Heart Study, which were measured on plasma frozen at 2808C for 12 to 15 years.
To test the hypothesis that elevated levels of these three biomarkers associate with increased risk of comorbidities in COPD, we defined clinically useful cut-points for each biomarker. Although categorization may lead to loss of information compared with using the biomarkers as a continuous variable, we used these cut-points because they are simple and clinically useful. CRP levels were categorized using two cut-points, 1 and 3 mg/L, that previously have been used by us and others in cardiovascular/pulmonary medicine (4, 7, 15). We next defined equivalent cut-points for fibrinogen and leukocyte count: 9 and 14 mmol/L for fibrinogen and 6 and 9 3 109/L for leukocyte count, so that the numbers in each of the low groups for each biomarker were roughly the same. In the combined analyses, levels of CRP, fibrinogen, and leukocyte count were defined as high or low according to cut-points of 3 mg/L for CRP, 14 mmol/L for fibrinogen, and 9 3 109/L for leukocyte count. If the three biomarkers were divided into tertiles or quintiles, results were largely similar to those presented. Leukocyte count was not determined in the 1991 to 1994 examination of the Copenhagen City Heart Study, explaining the lower number of individuals included in these analyses.
On a subset of individuals (n ¼ 430) from the Copenhagen City Heart Study 1991 to 1994 and 2001 to 2003 examinations, we had replicate measurements of CRP and fibrinogen. The stability of these biomarkers over this 10-year period is shown in Figure E1. Due to combined effects of measurement errors, long-term fluctuations, and changes within persons, these values tend to attenuate toward the actual mean over time, and this may underestimate true associations (16). Risk estimates can be corrected for this regression dilution bias when examining biomarkers separately. However, it is not possible to apply this correction when biomarkers are analyzed in combination, and we have therefore not corrected for this in the presented analyses.
Endpoints
Information on diagnosis of ischemic heart disease,myocardial infarction, heart failure, type II diabetes, lung cancer, pneumonia, pulmonary embolism, hip fracture, and depression was collected by reviewing all hospital admissions and diagnoses entered in the national Danish PatientRegistry, the national Danish Cancer Registry, and the national Danish Causes of Death Registry. Records include admission and/or death date and diagnoses according to the World Health Organization International Classification of Diseases (ICD8 or ICD10). All individuals with records of ischemic heart disease (ICD8: 410–414; ICD10: I20–I25), myocardial infarction (ICD8: 410; ICD10: I21–I22), heart failure (ICD8: 427.09– 427.11; ICD10: I50), type II diabetes (ICD8: 250; ICD10: E11, E13, E14), lung cancer (ICD10: C34), pneumonia (ICD8: 480–486; ICD10: J12–J18), pulmonary embolism (ICD8: 450.99, 673.99; ICD10: I26.0, I26.9, O88.2), hip fracture (ICD8: 820; ICD10: S720–S722), and depression (ICD8: 296.0, 296.2, 298.0, 300.4; ICD10: F32–F33) from study entry to end of follow-up in May 2011 were considered as having an event. If an individual had multiple records of one endpoint, the first occurrence was considered the event. The data obtained from the national Danish Causes of Death Registry report death from all comorbidities; the National Danish Cancer Registry describes death/hospitalization from cancer comorbidities. However, data from the Danish Patient Registry combine hospitalization from or with comorbidities.
Covariates
Participants were categorized as current smokers, former smokers, or never smokers. Cumulative tobacco consumption was calculated in pack-years, defined as 20 cigarettes/d/yr or equivalent. Body mass index was calculated asmeasuredweight (kilograms) divided bymeasured height squared (meters squared). Hypertension was self-reported use of antihypertensive medication. Plasma levels of total cholesterol, triglycerides, and high-density lipoprotein cholesterolweremeasured using standard hospital assays. A COPD exacerbation was defined as hospitalization from COPD (ICD8: 490–492; ICD10: J44) and/or respiratory failure (ICD10: J96) in patients with COPD. Exacerbation rate per year was number of exacerbations for each individual divided by years of follow-up.
Statistical Analysis
Statistical analyses were performed using STATA/SE version 11.1. We used a Cox proportional hazards regression model with age as time scale to estimate hazard ratios with 95% confidence intervals (CIs). The proportional hazard assumption was judged by visual inspection of cumulative hazard logarithm plots against age; no major violations were observed. For each endpoint, individuals with events before study entry were excluded. Models were adjusted either for age, sex, smoking status, and cumulative tobacco consumption or multifactorially for age, sex, smoking status, cumulative tobacco consumption, body mass index, hypertension, cholesterol, triglycerides, and high-density lipoprotein cholesterol. The first model was included to avoid overadjustment of estimates for endpoints such as lung cancer and pneumonia, where conventional cardiovascular risk factors included in the multifactorially adjusted model may be of limited importance. Continuous covariates were grouped into deciles before analyses. For test for trend of risk estimates, groups based on increasing levels of CRP, fibrinogen, and/or leukocyte count were coded 1, 2, 3, etc. Absolute 5-year risk by groups of the three biomarkers was estimated using the regression coefficients from a Poisson regression model (17).
Numbers of individuals with missing values for covariates are seen in Table E2. For continuous covariates, missing values were imputed using linear regression analysis with age and sex as predictors. If analyses were limited to individuals withoutmissing values, results were similar to those presented.
Results
Baseline characteristics of the 8,656 participants with COPDidentified by FEV1/FVC less than 0.7 with measurements of all three biomarkers are shown in Table 1. As expected, individuals with GOLD 3 to 4 were older, more likely to be male and current smokers, had higher cumulative tobacco consumption, and had higher levels of inflammatory biomarkers than individuals with GOLD 1 to 2. Defining COPD by FEV1/FVC below the lower limit of normal identified 6,555 individuals, who were younger, more likely to be current smokers, and had lower FEV1 % predicted than individuals identified by FEV1/FVC less than 0.7 (Table E3). During a median of 5 years’ follow-up of the 8,656 participants with COPD, 368 individuals were diagnosed with ischemic heart disease, 179 with myocardial infarction, 292 with heart failure, 143 with type II diabetes, 93 with lung cancer, 657 with pneumonia, 92 with pulmonary embolism, 141 with hip fracture, and 84 with depression. Number of individuals with diagnosed comorbidities before study entry is seen in Table 2. Characteristics of participants according to groups of CRP, fibrinogen, and leukocyte count are shown in Tables E4 to E6.
One Inflammatory Biomarker
Risk of ischemic heart disease, myocardial infarction, heart failure, type II diabetes, lung cancer, and pneumonia was increased by factors of 1.69 to 1.94 in individuals with CRP greater than 3 mg/L versus individuals with CRP less than 1 mg/L in multifactorially adjusted models (P for trend < 0.006) (Figure E2). There were no differences in risk of pulmonary embolism, hip fracture, and depression by levels of CRP (P for trend, 0.09– 0.86). Slightly different results were seen for fibrinogen and leukocyte count (Figures E3 and E4).
Three Inflammatory Biomarkers
To see whether high levels of three biomarkers in combination aided in identifying individuals at increased risk of comorbidities, we divided individuals into groups according to combinations of CRP, fibrinogen, and leukocyte count. Risk of ischemic heart disease was 2.19 (95% CI, 1.48–3.23) in individuals with three biomarkers elevated (CRP . 3 mg/L, fibrinogen . 14 mmol/L, and leukocyte count . 9 3 109/L) compared with individuals with all three biomarkers at or below these limits in multifactorially adjusted models (Figure 1). Corresponding hazard ratios were 2.32 (1.34–4.04) for myocardial infarction, 2.63 (1.71–4.04) for heart failure, 3.54 (2.03–6.19) for diabetes, 4.00 (2.12–7.54) for lung cancer, and 2.71 (2.03–3.63) for pneumonia (P for trend < 0.002). There were no overall differences in risk of pulmonary embolism, hip fracture, and depression by combinations of the three biomarkers (P for trend, 0.17–0.75); however, risk of hip fracture was 2.14 (1.19–3.85) in those with three high biomarkers versus individuals with three low biomarkers. For each endpoint, individuals with an event before study entry were excluded. If individuals with an event in the first year of follow-up were excluded, results were similar to those presented (Figure E5). Also, exclusion of individuals with a record of any of the nine endpoints before study entry (n ¼ 1,898) did not change main results (Figure E6).
Individuals with three high biomarkers had increased rates of severe exacerbations per year during follow-up compared with individuals with three low biomarkers (Table E7). If we excluded individuals with exacerbations in the year before date of examination (data not shown), or adjusted for exacerbation rate during follow-up, results were similar to those presented (Figure E7).
In stratified analyses, risk estimates for ischemic heart disease, myocardial infarction, heart failure, type II diabetes, and pneumonia in individuals with GOLD 1 to 2 remained statistically significantly increased and were similar to those presented in Figure 1 (Figure 2). A similar trend was observed for ischemic heart disease, myocardial infarction, heart failure, type II diabetes, and pneumonia in individuals with GOLD 3 to 4; however, not all risk estimates were statistically significant, and confidence intervals were wider due to the reduced statistical power. For lung cancer, risk was increased a factor of 5.76 (95% CI, 2.76–12.02), whereas there were no significant differences in GOLD 3 to 4.
In sensitivity analyses using FEV1/FVC below the lower limit of normal as COPD definition, risk estimates were similar to those presented in Figure 1 (Figure E8). Also, stratifying for smoking status showed similar results (Figure E9).
Absolute Risk
There was a stepwise increase in the absolute 5-year risk of ischemic heart disease, myocardial infarction, heart failure, type II diabetes, lung cancer, and pneumonia across groups of the three biomarkers combined in never smokers, formers smokers, and current smokers (Figure 3). In former smokers, the absolute 5-year risk for ischemic heart disease was 14% in individuals with three high biomarkers and 5% for individuals with three low biomarkers. Corresponding numbers were 7% and 2% for myocardial infarction, 11% and 4% for heart failure, 8% and 1% for type II diabetes, 5% and 1% for lung cancer, and 21% and 8% for pneumonia. In current smokers, numbers were similar but attenuated for all endpoints except lung cancer, where the absolute 5-year risk was 7% in individuals with three high biomarkers and 2% for individuals with three low biomarkers. Similar results were seen across age groups and body mass index categories (Figures E10 and E11).
Discussion
This study examined baseline measurements of CRP, fibrinogen, and leukocyte count in 8,656 individuals with COPD from two large population studies. We found that increased levels of these three inflammatory biomarkers in combination were associated with increased risk of ischemic heart disease, myocardial infarction, heart failure, type II diabetes, lung cancer, and pneumonia in individuals with COPD. Risk estimates were largely independent of smoking and ranged from a 2.2-fold increased risk of ischemic heart disease to a 4.0-fold increased risk of lung cancer when comparing individuals with three high biomarkers versus low values of these three biomarkers. These findings are novel.
The association between COPD and its comorbidities may be caused by common risk factors, such as smoking. However, studies taking smoking and similar factors into account still find a robust association (18). Several studies have suggested that common inflammatory pathways might be the link between COPD and its comorbidities (6, 18, 19), although the exact biological mechanisms remain unclear. In support of this hypothesis, studies from the non-COPD population show that high levels of CRP associate with increased risk of ischemic heart disease (15), cancer (20), and depression (21). However, it seems that for some diseases (e.g., depression) associations apparent in the non-COPD population cannot be reproduced in our COPD population. One explanation could be that the state of low-grade inflammation in patients with COPD due to pulmonary manifestations alone masks the associations otherwise found in healthy individuals. In any case, markers of systemic inflammation seem to be valuable in identifying patients with COPD at increased risk of cardiovascular disease, type II diabetes, lung cancer, and pneumonia, as seen in our study.
We found similar associations according to GOLDgrades for all endpoints except lung cancer, suggesting that findings are independent of lung function. In line with this, a recent study found that persistent systemic inflammation in patients with COPD associate with increased all-causemortality and exacerbation frequency compared with noninflamed patients despite similar lung function impairment in the two groups (22). The lack of significant association for lung cancer in GOLD 3 to 4 and the wider confidence intervals in general in GOLD 3 to 4 are most likely due to lower power. Also, frequent exacerbations, infections, and medication in these individuals (23) may affect levels of inflammatory biomarkers (6, 24); thus, associations apparent in GOLD 1 to 2 may not be seen in GOLD 3 to 4 due to much lower number of individuals in the latter subgroup. Elevated levels of CRP have previously been associated with increased risk of pulmonary embolism, hip fracture, and depression in general population studies (21, 25–27), but we were not able to find these associations in our COPD cohort. One explanation could be that the degree of systemic inflammation in COPD blurs these associations or that inflammation perhaps plays a limited role in the pathophysiology for these diseases.
Some limitations of our study must be considered in evaluating our results. Despite including several important potential confounders in the Cox regression model, such as cumulative tobacco consumption and smoking status, we cannot exclude that our results might be influenced by residual confounding. Another limitation is that we only had prebronchodilator measurements available. Also, recruitment of patients with COPD from the general population could lead to selection bias due to possible overrepresentation of relatively healthy patients with COPD, but this would tend to draw the results in a direction toward the null hypothesis and cannot explain our positive results. Another limitation is that our results have not been validated externally.
Also, we were not able to correct for regression dilution bias, since this correction is based on examining biomarkers separately (16); however, lack of such correction will only tend to underestimate real risk associations. We have not corrected P values for multiple comparisons, but all major findings remained statistically significant if results were Bonferroni corrected for the tests performed on the nine parallel endpoints.
Comorbidities contribute to poor health outcome in COPD, with cardiovascular disease and lung cancer being leading causes of morbidity and mortality in patients with COPD (12, 28–30). Thus, early detection and treatment of comorbidities will have beneficial effects on the clinical course of individuals with COPD. However, due to an extensive overlap of symptoms (e.g., with dyspnea in heart failure and weight loss in lung cancer), it may be difficult to distinguish between symptoms related to comorbidities from those related to COPD itself. Therefore, measurements of CRP, fibrinogen, and leukocyte count in patients with COPD may, in addition to the usual clinical examination and assessment of traditional risk factors, help select individuals with increased risk and need of additional diagnostic evaluation. This is especially relevant for cardiovascular disease and type II diabetes, where prophylactic approaches are feasible and effective, and for lung cancer, where early detection by computed tomography screening will improve prognosis (31).
In conclusion, simultaneously elevated levels of CRP, fibrinogen, and leukocyte count are associated with a two- to fourfold risk of major comorbidities in COPD. These biomarkers may be an additional tool for clinicians to conduct stratified management of comorbidities in patients with COPD.
Author disclosures are available with the text of this article at www.atsjournals.org.
References
Влезте или се регистрирайте безплатно, за да получите достъп до пълното съдържание и статиите на списанието в PDF формат.1. Rabe KF, Hurd S, Anzueto A, Barnes PJ, Buist SA, Calverley P, Fukuchi Y, Jenkins C, Rodriguez-Roisin R, van Weel C, et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med 2007;176:532–555.
2. Dahl M, Tybjaerg-Hansen A, Vestbo J, Lange P, Nordestgaard BG. Elevated plasma fibrinogen associated with reduced pulmonary function and increased risk of chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2001;164:1008–1011.
3. Gan WQ, Man SF, Senthilselvan A, Sin DD. Association between chronic obstructive pulmonary disease and systemic inflammation: a systematic review and a meta-analysis. Thorax 2004;59:574–580.
4. Dahl M, Vestbo J, Lange P, Bojesen SE, Tybjaerg-Hansen A, Nordestgaard BG. C-reactive protein as a predictor of prognosis in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2007;175:250–255.
5. Fabbri LM, Luppi F, Beghe B, Rabe KF. Complex chronic comorbidities of COPD. Eur Respir J 2008;31:204–212.
6. Barnes PJ, Celli BR. Systemic manifestations and comorbidities of COPD. Eur Respir J 2009;33:1165–1185.
7. Dahl M, Vestbo J, Zacho J, Lange P, Tybjaerg-Hansen A, Nordestgaard BG. C reactive protein and chronic obstructive pulmonary disease: a Mendelian randomisation approach. Thorax 2011;66:197–204.
8. Mannino DM, Brown C, Giovino GA. Obstructive lung disease deaths in the United States from 1979 through 1993. An analysis using multiplecause mortality data. Am J Respir Crit Care Med 1997;156:814–818.
9. Van Eeden S, Leipsic J, Man SF, Sin DD. The relationship between lung inflammation and cardiovascular disease. Am J Respir Crit Care Med 2012;186:11–16.
10. Sin DD, Man JP, Man SF. The risk of osteoporosis in Caucasian men and women with obstructive airways disease. Am J Med 2003;114:10–14.
11. Holguin F, Folch E, Redd SC, Mannino DM. Comorbidity and mortality in COPD-related hospitalizations in the United States, 1979 to 2001. Chest 2005;128:2005–2011.
12. Sode BF, Dahl M, Nordestgaard BG. Myocardial infarction and other co-morbidities in patients with chronic obstructive pulmonary disease: a Danish nationwide study of 7.4 million individuals. Eur Heart J 2011;32:2365–2375.
13. Nordestgaard BG, Benn M, Schnohr P, Tybjaerg-Hansen A. Nonfasting triglycerides and risk of myocardial infarction, ischemic heart disease, and death in men and women. JAMA 2007;298:299–308.
14. Frikke-Schmidt R, Nordestgaard BG, Stene MC, Sethi AA, Remaley AT, Schnohr P, Grande P, Tybjaerg-Hansen A. Association of lossof- function mutations in the ABCA1 gene with high-density lipoprotein cholesterol levels and risk of ischemic heart disease. JAMA 2008;299:2524–2532.
15. Zacho J, Tybjaerg-HansenA, Jensen JS, Grande P, Sillesen H, Nordestgaard BG. Genetically elevated C-reactive protein and ischemic vascular disease. N Engl JMed 2008;359:1897–1908.
16. Clarke R, Shipley M, Lewington S, Youngman L, Collins R, Marmot M, Peto R. Underestimation of risk associations due to regression dilution in longterm follow-up of prospective studies. Am J Epidemiol 1999;150:341–353.
17. Juul K, Tybjaerg-Hansen A, Schnohr P, Nordestgaard BG, Factor V. Leiden and the risk for venous thromboembolism in the adult Danish population. Ann Intern Med 2004;140:330–337.
18. Sevenoaks MJ, Stockley RA. Chronic Obstructive Pulmonary Disease, inflammation and co-morbidity–a common inflammatory phenotype? Respir Res 2006;7:70.
19. Fabbri LM, Rabe KF. From COPD to chronic systemic inflammatory syndrome? Lancet 2007;370:797–799.
20. Allin KH, Bojesen SE, Nordestgaard BG. Baseline C-reactive protein is associated with incident cancer and survival in patients with cancer. J Clin Oncol 2009;27:2217–2224.
21. Wium-Andersen MK, Ørsted DD, Nielsen SF, Nordestgaard BG. Elevated C-reactive protein, psychological distress, and depression in 73,131 individuals. Arch Gen Psychiatry (In press)
22. Agusti A, Edwards LD, Rennard SI, Macnee W, Tal-Singer R, Miller BE, Vestbo J, Lomas DA, Calverley PM, Wouters E, et al. Persistent systemic inflammation is associated with poor clinical outcomes in COPD: a novel phenotype. PLoS ONE 2012;7:e37483.
23. Hurst JR, Vestbo J, Anzueto A, Locantore N, Mullerova H, Tal-Singer R, Miller B, Lomas DA, Agusti A, Macnee W, et al. Susceptibility to exacerbation in chronic obstructive pulmonary disease. N Engl J Med 2010;363:1128–1138.
24. Groenewegen KH, Postma DS, Hop WC, Wielders PL, Schlosser NJ, Wouters EF. Increased systemic inflammation is a risk factor for COPD exacerbations. Chest 2008;133:350–357.
25. Vormittag R, Vukovich T, Schonauer V, Lehr S, Minar E, Bialonczyk C, Hirschl M, Pabinger I. Basal high-sensitivity-C-reactive protein levels in patients with spontaneous venous thromboembolism. Thromb Haemost 2005;93:488–493.
26. Cauley JA, Danielson ME, Boudreau RM, Forrest KY, Zmuda JM, Pahor M, Tylavsky FA, Cummings SR, Harris TB, Newman AB. Inflammatory markers and incident fracture risk in older men and women: the Health Aging and Body Composition Study. J BoneMiner Res 2007;22: 1088–1095.
27. Howren MB, Lamkin DM, Suls J. Associations of depression with C-reactive protein, IL-1, and IL-6: a meta-analysis. Psychosom Med 2009;71:171–186.
28. Anthonisen NR, Skeans MA, Wise RA, Manfreda J, Kanner RE, Connett JE. The effects of a smoking cessation intervention on 14.5-year mortality: a randomized clinical trial. Ann Intern Med 2005;142:233–239.
29. Mannino DM, Doherty DE, Sonia BA. Global Initiative on Obstructive Lung Disease (GOLD) classification of lung disease and mortality: findings from the Atherosclerosis Risk in Communities (ARIC) study. Respir Med 2006;100:115–122.
30. Calverley PM, Anderson JA, Celli B, Ferguson GT, Jenkins C, Jones PW, Yates JC, Vestbo J. Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease. N Engl J Med 2007; 356:775–789.
31. Aberle DR, Adams AM, Berg CD, Black WC, Clapp JD, Fagerstrom RM, Gareen IF, Gatsonis C, Marcus PM, Sicks JD. Reduced lungcancer mortality with low-dose computed tomographic screening. N Engl J Med 2011;365:395–409.