Systemic Inflammation and COPD*

Transcription

1 CHEST Systemic Inflammation and COPD* The Framingham Heart Study Original Research Robert E. Walter, MD, MPH; Jemma B. Wilk, DSc; Martin G. Larson, ScD; Ramachandran S. Vasan, MD; John F. Keaney, Jr, MD; Izabella Lipinska, PhD; George T. O Connor, MD, MS, FCCP; and Emelia J. Benjamin, MD, ScM COPD Background: The current paradigm for the pathogenesis of COPD includes an ultimately maladaptive local inflammatory response to environmental stimuli. We examined the hypothesis that systemic inflammatory biomarkers are associated with impaired lung function, particularly among those with extensive cigarette smoking. Methods: Using data from the Framingham Heart Study, we examined cross-sectional associations of systemic inflammatory biomarkers (CD40 ligand [CD40L], intercellular adhesion molecule [ICAM]-1, interleukin [IL]-6, monocyte chemoattractant protein-1, P-selectin, and myeloperoxidase, in addition to C-reactive protein) to impaired lung function. Results: IL-6 was consistently associated with impaired lung function; a 1-SD higher concentration of IL-6 was associated with a 41-mL lower FEV 1 (95% confidence interval [CI], 61 to 20) and a borderline 15% higher odds of COPD (odds ratio, 1.15; 95% CI, 0.99 to 1.34). Additionally, P-selectin was associated with lower FEV 1 levels; after adjusting for the other biomarkers, a 1-SD higher concentration of P-selectin predicted an FEV 1 that was on average 19 ml lower (95% CI, 37 to 0). Including the biomarkers individually as sole exposures in the models generally strengthened the impaired lung function/biomarker association; the relations of ICAM-1 to FEV 1, and ICAM and CD40L to COPD became significant. The observed associations did not vary significantly with smoking history, except that the association between CD40L and COPD appeared greater in individuals with more extensive smoking histories. Conclusions: Among participants in the Framingham Heart Study, systemic inflammation was associated with lower levels of pulmonary function. Further research into the role of systemic inflammation in the development of pulmonary dysfunction is merited. (CHEST 2008; 133:19 25) Key words: CD40 ligand; forced expiratory volume; inflammation; intercellular adhesion molecule-1; interleukin-6; monocyte chemoattractant protein-1; myeloperoxidase; p-selectin; pulmonary disease, chronic obstructive Abbreviations: CD40L CD40 ligand; CI confidence interval; CRP C-reactive protein; ICAM intercellular adhesion molecule; IL interleukin; MCP monocyte chemoattractant protein; MPO myeloperoxidase; OR odds ratio The current paradigm for the pathogenesis of COPD includes a local inflammatory response to environmental challenges, most notably cigarette smoking, which ultimately results in the development of airflow obstruction. 1 Signaled by a variety of cytokines secreted in response to the initial insult, inflammatory cells, especially macrophages, neutrophils, and T-lymphocytes, are recruited into the airways and activated, further propagating the inflammatory cascade. These processes add to the environmental oxidant load, overwhelming the intrinsic antioxidant potential. In the susceptible host, the combination of local inflammation and oxidant injury leads to small airways narrowing and fibrosis, which become clinically manifest as COPD. Serum concentrations of C-reactive protein (CRP), a biomarker of systemic inflammation, have been associ- For editorial comment see page 4 ated with impaired lung function. The interaction between local and systemic inflammation in the development of chronic airflow obstruction is of CHEST / 133 / 1/ JANUARY,

2 interest, although its nature not well understood. Higher levels of systemic inflammation may simply represent spill over from airways inflammation, or the initial local response may be modified by systemic factors. The specific role of CRP in inflammatory pathways is not completely understood, which limits any inferences that can be made from its association with impaired lung function as to the pathogenesis of chronic airflow obstruction in the susceptible host. Other biomarkers known to influence the mechanisms implicated in development of COPD may offer additional avenues of research. There are, however, only a limited number of reports to date describing associations between either systemic or local concentrations of other biomarkers to impaired lung function, and these are generally based on small case-control studies. Understanding the relation of systemic concentrations of such biomarkers of inflammation and oxidant stress to impaired lung function may provide insights into intrapulmonary processes leading to chronic airflow obstruction in the susceptible host, and a better understanding of the relation of systemic inflammation to those intrapulmonary processes. The current paradigm for the pathogenesis of COPD offers a number of credible and biologically informative biomarkers. 2 Monocytes, prominent in the histopathology of COPD, are activated and recruited by both intercellular adhesion molecule (ICAM)-1 and monocyte chemoattractant protein (MCP)-1. P-selectin has been described to promote neutrophil recruitment into the lung in models of lung injury. 3 CD40 ligand (CD40L), found both on activated T cells and in soluble form, contributes to the activation of both B and T lymphocytes, as well as macrophages, while interleukin (IL)-6 induces T-cell differentiation. Myeloperoxidase (MPO), an enzyme *From the Section of Pulmonary, Allergy, and Critical Care Medicine (Drs. Walter and O Connor), Department of Medicine (Dr. Wilk), and Section of Pulmonary and Critical Care Medicine (Drs. Vasan and Benjamin), Boston University School of Medicine, Boston; Department of Mathematics and Statistics (Dr. Larson), Boston University, Boston; Division of Cardiovascular Medicine (Dr. Keaney), University of Massachusetts Medical School, Worcester; and Whitaker Cardiovascular Institute (Dr. Lipinska), Boston, MA. Support was provided by National Institutes of Health/National Heart, Lung, and Blood Institute contracts N01-HC-25195, N01-HV-28178, HL076784, HL064753, and AG Drs. Walter and Wilk are each recipients of a Young Clinical Scientist Award from the Flight Attendant Medical Research Institute. The authors have no conflicts of interests to disclose. Manuscript received January 8, 2007; revision accepted July 22, Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( org/misc/reprints.shtml). Correspondence to: Robert E. Walter, MD, MPH, The Pulmonary Center, 80 East Concord, R304, Boston, MA 02118; walterb@bu.edu DOI: /chest found both in monocytes and especially neutrophils, aids in bactericidal activities while contributing to the oxidant burden. We hypothesized that systemic inflammation contributes to the intraparenchymal processes resulting in chronic airflow obstruction, and that the systemic inflammatory profile associated with impaired lung function in heavy smokers differs from that associated with airflow obstruction in those with little or no cigarette use. Using the Framingham Heart Study, we tested the specific hypotheses that higher concentrations of these biomarkers were associated with lower levels of lung function and with the presence of COPD, and that the relation between the biomarkers and impaired lung function was modified by long-term smoking. Materials and Methods Subjects, Covariates, and Measures of Pulmonary Function and Systemic Inflammation Participants attending the most recent completed examination (ie, seventh, from 1998 to 2001) of the Offspring Cohort 4 were eligible for inclusion. Informed consent was obtained in accordance with the protocol approved by the Institutional Review Board of Boston University Medical Center. The single-day examination included a variety of procedures, with phlebotomy after an overnight fast (typically between 8:00 am and 9:00 am) and subsequent spirometry without bronchodilator testing performed according to American Thoracic Society standards. 5 Commercially available enzyme-linked immunosorbent assays were used to measure in duplicate serum concentrations of IL-6, ICAM-1, MCP-1, CD40L, and plasma concentrations of P- selectin and MPO. 6,7 The intraassay coefficients of variation were as follows: CD40L, 4.4%; IL-6, 3.1%; ICAM-1, 3.1%; MCP, 4.1%; MPO, 3.0%; and P-selectin, 3.0%. Serum CRP was measured once with a nephelometer (BN100; Dade Behring; Deerfield, IL). 8 The concentrations of biomarkers were log transformed to a more normal distribution. Statistical Analysis Using data for healthy never smokers, we developed sexspecific prediction models by regressing FEV 1 onto age, age squared, and height squared as described by Hankinson et al. 9 Predicted values were derived for all participants using the regression coefficients obtained, and residual values were calculated by subtracting the predicted value from the observed value. COPD was defined as an FEV 1 /FVC ratio 0.70 with FEV 1 80% of predicted, consistent with Global Initiative for Chronic Obstructive Lung Disease stage 2 or higher. 10 We examined the relation of the biomarkers to residual FEV 1 using linear regression to adjust for age, sex, body mass index, current smoking, and pack-years in all models. The primary model included all biomarkers (ICAM, IL-6, MCP, P-selectin, CD40L, and MPO), along with CRP concentrations, as exposures. There was, however, a statistically significant correlation among the biomarkers. CRP was especially correlated to the other markers, with its strongest correlation being to IL-6 (Spearman correlation coefficient, 0.49; p ). 6 To further characterize the degree of collinearity, we calculated the variance 20 Original Research

3 inflation factor by first regressing each biomarker onto the others, then dividing 1 by 1 R 2 from this regression (ie, variance inflation factor for biomarker i 1/[1 R 2 i]). Given the correlations of the biomarkers, simultaneous inclusion of the entire panel may mask the relation of individual biomarkers to impaired lung function; secondary analyses were implemented including only single biomarkers as exposures. Subjects were subsequently classified as those with heavy cigarette consumption (at least 10 pack-years of personal cigarette smoking 11,12 ) and those without such long-term exposure; the models were implemented among these subgroups. Subjects with complete information for all covariates in the model were included in that analysis. Using a similar approach, we analyzed the relation of biomarkers to COPD using logistic regression to adjust for age, sex, current smoking, and pack-years. The primary model included all markers as exposures; secondary models included the biomarkers individually. These models were implemented among the total sample and among the two subgroups. There were fewer cases of COPD among those without heavy smoking, and only one biomarker (IL-6) was significantly related to COPD; as a result, these analyses are not presented. Current smoking may confound the relation between the biomarkers and impaired lung function. As a result, we repeated the analyses above with both COPD and FEV 1 as dependent variables among all subjects not reporting current smoking, and among subjects with at least 10 pack-years of cigarette smoking but reporting they did not currently smoke. Exclusion of current smokers resulted in the loss of 260 participants with 73 cases of COPD from the overall cohort, and exclusion of current smokers from the subgroup with at least 10 pack-years of smoking resulted in the loss of 144 participants and 70 cases of COPD. While the strength of the associations of the biomarkers to impaired lung function was weakened, the effect size was not substantively changed. As a result, these analyses are not presented. To test the significance of apparent differences between subgroups of cigarette smoking in the associations of biomarker concentration to impaired lung function (ie, IL-6, P-selectin, and CRP in the association with FEV 1, and CD40L with COPD), we implemented models among all participants that included baseline covariates, smoking history as a dichotomous variable ( 10 pack-years, yes/no), the biomarker of interest, and the interaction term for the biomarker and smoking strata. All statistical analyses were performed using software (version 9.1; SAS Institute; Cary, NC). A two-sided p value 0.05 was considered statistically significant. Results Subject Characteristics Participant characteristics are described in Table 1; the majority were female and had a history of cigarette smoking. Compared to those without evidence of chronic airflow obstruction, those with COPD were older, were more likely to be smokers, and had more pack-years of smoking exposure. Such differences between subjects with and without COPD also existed among the subgroup with at least 10 pack-years of cigarette smoking. Additionally, subjects with COPD generally had higher concentrations of circulating serum inflammatory markers, including CRP. There was, however, only modest correlation among the biomarkers (Table 2); the variance inflation factors ranged only between and There was a range of severity of COPD in our community-based cohort; among all participants with COPD regardless of smoking history, the median FEV 1 was 69% of predicted and the lowest Table 1 Participant Characteristics* All Participants Subgroup With 10 Pack-Years Characteristics COPD (n 309) No COPD (n 2,244) COPD (n 206) No COPD (n 773) Clinical characteristics Age, yr 63.7 (8.8) 60.0 (9.3) 63.0 (8.1) 60.7 (9.0) Male gender Height, inches 65.8 (3.8) 66.0 (3.7) 66.0 (3.7) 66.6 (3.5) Body mass index, kg/m (5.7) 28.1 (5.2) 28.6 (4.9) 28.6 (5.5) FEV 1, % predicted 66.1 (11.4) 99.0 (13.4) 64.7 (11.6) 96.1 (12.7) FEV 1 /FVC 0.59 (0.08) 0.74 (0.05) 0.58 (0.08) 0.73 (0.06) Smoking history Former smoker Pack-yr 37.6 (28.2) 19.8 (20.9) 42.8 (26.5) 30.0 (21.0) Current smoker Pack-yr 39.9 (21.5) 27.0 (39.9) 43.4 (19.4) 33.4 (20.2) Biomarkers CD40L, ng/ml 3.9 (3.6) 3.8 (3.4) 4.3 (3.8) 4.0 (3.6) ICAM-1, ng/ml 278 (93) 251 (75) 289 (95) 263 (79) IL-6, pg/ml 4.9 (4.8) 3.7 (5.0) 5.3 (5.4) 4.3 (5.8) MCP-1, pg/ml 347 (130) 321 (123) 350 (107) 335 (150) P-selectin, ng/ml 40.7 (14.0) 37.1 (13.6) 42.2 (15.0) 38.7 (14.1) MPO, ng/ml 48.9 (30.4) 47.1 (29.4) 51.3 (33.0) 48.0 (29.5) CRP, mg/l 5.8 (8.3) 3.9 (7.6) 6.5 (9.5) 4.5 (10.8) *Data are presented as mean (SD) or %. CHEST / 133 / 1/ JANUARY,

4 Table 2 Collinearity of the Biomarkers* Variables R 2 Variance Inflation Factor CD40L ICAM IL MCP P-selectin MPO CRP *Values are the R 2 from regression of each biomarker onto the others, and the variance inflation factor calculated from that R 2 (variance inflation factor for biomarker i 1/ 1 R 2 i ). quartile ranged from 24 to 60% of predicted. Among the heavier smokers with COPD, the median FEV 1 was 66% of predicted and the lowest quartile ranged from 24 to 58% of predicted. Relation of Circulating Markers of Inflammation and Oxidant Stress to FEV 1 Using linear regression adjusting for age, sex, body mass index, current smoking, and pack-years, we examined the relation of the biomarkers to residual FEV 1. With the full panel of biomarkers included as exposures (Table 3), IL-6 and P-selectin were significantly associated with FEV 1 among all subjects. After adjusting for potential confounders and the other biomarkers, 1-SD higher concentrations of IL-6 and P-selectin were associated with an FEV 1 that was on average 41 ml lower (95% confidence interval [CI], 61 to 20) and 19 ml lower (95% CI, 37 to 0), respectively. Additionally, higher CRP concentrations were associated with lower levels of FEV 1 ; a 1-SD higher concentration of CRP was associated with an FEV 1 that was 46 ml lower (95% CI, 67 to 24). Repeating the analyses among all participants with the biomarkers individually included as exposures (Table 3) generally strengthened the associations of IL-6 and P-selectin to impaired lung function, and the relation of ICAM to impaired lung function became statistically significant. With ICAM as the only biomarker in the model, a 1-SD higher level predicted an FEV 1 that was 34 ml lower on average (95% CI, 52 to 16). Relation of Circulating Markers of Inflammation and Oxidant Stress to COPD Using logistic regression to adjust for age, sex, current smoking, and pack-years, we examined the relation of the biomarkers to COPD. Among all participants, only the relation of IL-6 concentrations to COPD approached significance (Table 4); a 1-SD higher concentration of IL-6 tended to be associated with a 15%-higher odds of COPD (odds ratio [OR], 1.15; 95% CI, 0.99 to 1.34). CRP concentration was significantly associated with the presence of COPD if all the biomarkers were included as exposures; a 1-SD higher CRP concentration was associated with 20%-higher odds of COPD (OR, 1.20; 95% CI, 1.02 to 1.40). Repeating the analyses with the biomarkers in individual models (Table 4) strengthened the association of IL-6 to impaired lung function. A 1-SD higher concentration of IL-6 was associated with 27%-higher odds of COPD (OR, 1.27; 95% CI, 1.12 to 1.44). While the relations of the other biomarkers to impaired lung function were some- Table 3 Relations of Biomarkers of Inflammation to FEV 1 * Variables All Participants (n 2,553) Subjects Without at Least 10 Pack-Years (n 1,574) Smokers With at Least 10 Pack-Years (n 979) All biomarkers included in one model, ml IL-6 41 ( 61, 20) 55 ( 78, 29) 17 ( 55, 20) P-selectin 19 ( 37, 0) 14 ( 37, 9) 28 ( 61, 4) ICAM 17 ( 37, 2) 17 ( 41, 8) 19 ( 53, 15) CRP 46 ( 67, 24) 34 ( 60, 8) 63 ( 101, 26) Biomarkers examined in separate models, ml IL-6 63 ( 81, 45) 70 ( 93, 47) 57 ( 88, 25) P-selectin 28 ( 45, 10) 20 ( 41, 2) 40 ( 71, 9) CD40L 1 ( 17, 17) 20 ( 2, 41) 26 ( 56, 5) ICAM-1 34 ( 52, 16) 35 ( 58, 11) 35 ( 67, 3) CRP 64 ( 84, 45) 53 ( 77, 29) 79 ( 111, 47) *Results for any biomarker with p 10 in that analysis (ie, that row) are presented as the predicted difference in FEV 1 (95% confidence limits) associated with a 1-SD higher concentration of biomarker of interest, using linear regression models to adjust for age, gender, body mass index, current smoking, and pack-years. p Original Research

5 Table 4 Relations of Biomarkers of Inflammation and Oxidant Stress to COPD* Variables All Participants (n 2,553) Smokers With at Least 10 Pack-Years (n 979) All biomarkers included in one model IL (0.99, 1.34) 1.08 (0.88, 1.33) CD40L 1.05 (0.90, 1.23) 1.20 (0.98, 1.46) CRP 1.20 (1.02, 1.40) 1.25 (1.03, 1.52) Biomarkers examined in separate models IL (1.12, 1.44) 1.27 (1.08, 1.50) P-selectin 1.11 (0.97, 1.27) 1.16 (0.98, ) CD40L 1.07 (0.93, 1.23) 1.21 (1.01, 1.45) ICAM (0.96, 1.24) 1.17 (1.0, 1.38) MPO 1.13 (0.89, 1.42) 1.16 (0.98, 1.36) CRP 1.31 (1.15, 1.50) 1.36 (1.15, 1.61) *Results for any biomarker with p 0.10 in that analysis (ie, that row) are presented as OR (95% CI) associated with a 1-SD higher level of the particular biomarker, using logistic regression to adjust for age, gender, current smoking, and pack-years. p what stronger when included as exposures individually, none of these relations achieved statistical significance. Similarly strengthened, a 1-SD higher concentration of CRP was associated with 31%-higher odds of COPD (OR, 1.31; 95% CI, 1.15 to 1.50). Relation of Biomarkers to Impaired Lung Function Among Smoking Strata Examining the relations of the individual biomarkers to FEV 1 among the two strata of smoking histories (Table 3) suggested some apparent differences between the groups. If the biomarkers were individually included as exposures, the relations of CRP and P-selectin to FEV 1 appeared stronger and IL-6 appeared weaker among the heavier vs lighter smokers; however, these apparent subgroup differences were not statistically significant (p values for biomarker smoking history interaction term 0.30 for CRP, 0.41 for IL-6, and 0.23 for P-selectin). Similarly, examining the relations of the individual biomarkers to COPD among the two strata of cigarette smoking (Table 4) suggested some differences between the groups. If the biomarkers were included individually as exposures among only those with more extensive smoking history, the relations of both CD40L and ICAM to COPD became statistically significant; these relations were not significant among those without at least 10 pack-years of cigarette smoking. The suggested differences between the smoking strata were only statistically significant for CD40L; compared to its relation among those without at least 10 pack-years of cigarette smoking (OR, 0.84; 95% CI, 0.68 to 1.05), the relation of CD40L to COPD among the heavier smokers was significantly stronger (p value for interaction term 0.01). Discussion We observed significant cross-sectional associations between a number of the biomarkers of systemic inflammation and impaired lung function. Although there appeared to be some differences in the associations with impaired lung function between the smoking strata, in our exploratory analyses these differences were statistically significant only for the relation of CD40L to COPD. CRP may have direct immunomodulatory effects in the lung and has been previously associated with impaired lung function. In clinic-based samples, higher concentrations of CRP have been associated with the presence of COPD 17 and CRP concentrations appear to decline with inhaled corticosteroid treatment for stable COPD. 18 Higher levels of CRP also have been associated with lower levels of FEV 1 in population studies In contrast, to our knowledge no communitybased studies have examined the relation of the other biomarkers in our panel to impaired lung function. In case-control studies, higher concentrations of IL-6, a cytokine that activates inflammatory cells, 22 have been observed in exhaled breath condensate of COPD subjects, compared to controls subjects. 23,24 Among participants in the seventh examination of the Framingham Heart Study, there was a consistent cross-sectional association of IL-6 to lower levels of lung function. ICAM influences intercellular interactions, especially in monocytes, which play a significant role in COPD. Although the relation of serum concentrations of ICAM to impaired lung function have not previously been described, bronchial epithelial cells from patients with COPD, compared to those from healthy smokers, secrete more ICAM 25 when exposed to cigarette smoke. Additionally, higher ICAM concentrations were seen in the airways of individuals with bronchitis with airflow obstruction, compared to those with normal airflow, 26 but sputum concentrations of ICAM did not significantly correlate with FEV Among our cohort, higher concentrations of serum ICAM, if the sole biomarker in the model, were associated with lower levels of FEV 1 in both strata of smoking history, and with a higher odds for COPD among those with extensive smoking histories. CHEST / 133 / 1/ JANUARY,

6 P-selectin has a role in platelet aggregation and influences neutrophil activation. In a case control study, Ferroni et al 28 reported higher concentrations of plasma P-selectin in COPD subjects compared to healthy control subjects, and that P-selectin concentrations were correlated with impaired gas exchange. In the Offspring Cohort, P-selectin levels were associated with lower levels of FEV 1 but its relation to COPD did not reach statistical significance. CD40L is a protein of the tumor necrosis factor family; its receptor is found on a variety of inflammatory cells and is known to influence the inflammatory state. 29 The relation, however, of CD40L concentration to impaired lung function has not previously been reported. In this cohort, we did not see a significant relation of CD40L to FEV 1, but we did observe an interaction with smoking status such that higher odds of COPD were observed with higher concentrations of CD40L among the heavy smokers. In contrast to previous reports, 30,31 we found no significant relation of impaired lung function to either MCP or MPO. MCP, known to play a role in monocyte recruitment, was seen in higher concentrations in the sputum and lavage fluid of smokers with airflow limitation, compared to that of smokers with normal lung function, and concentrations negatively correlated with FEV 1. MPO, an enzyme found in monocytes and especially neutrophils, aids in bactericidal activities while adding to the oxidant burden. Circulating concentrations of MPO have been reported to fall with treatment of COPD, 32 although any association with impaired lung function has not otherwise been described. We note several limitations of our study. We observed a cross-sectional relation of several circulating inflammatory biomarkers to impaired lung function; no cross-sectional study can definitively prove a causal link between the inflammatory cascades that incorporate these inflammatory markers and COPD. Additionally, as noted above, the relations between markers of inflammation and the tissue-level intrapulmonary processes that result in chronic airflow obstruction are not well understood; whether the systemic inflammation is an epiphenomenon or it contributes to the development of COPD is unknown. We used data from an ambulatory, community-based cohort; we acknowledge that the pathophysiology of the inflammation/ spirometry relation in individuals with advanced COPD, such as seen in a typical referral-based pulmonary clinic, may differ. Similarly, we assessed circulating inflammatory markers, which may not adequately reflect local pulmonary tissue inflammatory concentrations. While the differences between the smoking strata in associations of the inflammatory markers to impaired lung function are interesting, they may be explained by factors other than biological relevance and, in fact, the differences were not statistically significant except in the case of CD40L. A weaker association in one subgroup compared to the other could either be explained by the relative contribution of the inflammatory marker to the pathogenesis of airflow obstruction in the strata, or simply by a reduction in the power resulting from the smaller numbers. A stronger association in one subgroup vs the other would seem more likely to be of biological relevance; the difference, however, was of statistical significance only for the relation of CD40L to COPD. Ultimately, we did not observe large differences between the smoking strata. Our primary model would predict a current smoker of 20 pack-years duration would have an FEV 1 approximately 280 ml lower on average than a never-smoker of the same age, sex, and body mass index. By way of comparison, a 1-SD higher IL-6 level was associated, after adjustment for the other biomarkers, with an FEV 1 that was 41 ml lower, or approximately 15% of the association of current smoking with 20 pack-years, and the association of P-selectin to impaired lung function was approximately half that of IL-6. Although we cannot prove a causal relation, the strength of the association between these markers of systemic inflammation and impaired lung function would suggest that further investigation is warranted. An improved understanding of the role of systemic inflammation in the pathogenesis of COPD, especially as it relates to cigarette smoking, will help us better understand the different mechanisms that contribute to the COPD phenotype. These biomarkers CD40L, ICAM, and especially CRP, IL-6, and P-selectin offer promising avenues for research. References 1 MacNee W. Pathogenesis of chronic obstructive pulmonary disease. Proc Am Thorac Soc 2005; 2: Hogg JC, Chu F, Utokaparch S, et al. The nature of small-airway obstruction in chronic obstructive pulmonary disease. N Engl J Med 2004; 350: Mulligan MS, Polley MJ, Bayer RJ, et al. Neutrophil-dependent acute lung injury: requirement for P-selectin (GMP-140). J Clin Invest 1992; 90: Kannel WB, Feinleib M, McNamara PM, et al. An investigation of coronary heart disease in families: the Framingham offspring study. Am J Epidemiol 1979; 110: Standardization of spirometry, 1994 update. Am J Respir Crit Care Med 1995; 152: Dupuis J, Larson MG, Vasan RS, et al. Genome scan of systemic biomarkers of vascular inflammation in the Framingham Heart Study: evidence for susceptibility loci on 1q. Atherosclerosis 2005; 182: Loucks EB, Sullivan LM, Hayes LJ, et al. Association of educational level with inflammatory markers in the Framing- 24 Original Research

7 ham Offspring Study. Am J Epidemiol 2006; 163: Kathiresan S, Larson MG, Vasan RS, et al. Contribution of clinical correlates and 13 C-reactive protein gene polymorphisms to interindividual variability in serum C-reactive protein level. Circulation 2006; 113: Hankinson JL, Odencrantz JR, Fedan KB. Spirometric reference values from a sample of the general U.S. population. Am J Respir Crit Care Med 1999; 159: Pauwels RA, Buist AS, Calverley PM, et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: NHLBI/WHO Global Initiative for Chronic Obstructive Lung Disease (GOLD) Workshop summary. Am J Respir Crit Care Med 2001; 163: Hersh CP, DeMeo DL, Al Ansari E, et al. Predictors of survival in severe, early onset COPD. Chest 2004; 126: Zielinski J, Bednarek M, Know the Age of Your Lung Study Group. Early detection of COPD in a high-risk population using spirometric screening. Chest 2001; 119: Casals C, Arias-Diaz J, Valino F, et al. Surfactant strengthens the inhibitory effect of C-reactive protein on human lung macrophage cytokine release. Am J Physiol Lung Cell Mol Physiol 2003; 284:L466 L Heuertz RM, Ahmed N, Webster RO, et al. Peptides derived from C-reactive protein inhibit neutrophil alveolitis. J Immunol 1996; 156: Barna BP, Thomassen MJ, Zhou P, et al. Activation of alveolar macrophage TNF and MCP-1 expression in vivo by a synthetic peptide of C-reactive protein. J Leukoc Biol 1996; 59: Galve-de Rochemonteix B, Wiktorowicz K, Kushner I, et al. C-reactive protein increases production of IL-1, IL-1, and TNF-, and expression of mrna by human alveolar macrophages. J Leukoc Bio 1993; 53: Sin DD, Man SF. Why are patients with chronic obstructive pulmonary disease at increased risk of cardiovascular diseases? The potential role of systemic inflammation in chronic obstructive pulmonary disease. Circulation 2003; 107: Sin DD, Lacy P, York E, et al. Effects of fluticasone on systemic markers of inflammation in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2004; 170: Aronson D, Roterman I, Yigla M, et al. Inverse association between pulmonary function and C-reactive protein in apparently healthy subjects. Am J Respir Crit Care Med 2006; 174: Gan WQ, Man SF, Sin DD. The interactions between cigarette smoking and reduced lung function on systemic inflammation. Chest 2005; 127: Gan WQ, Man SF, Senthilselvan A, et al. Association between chronic obstructive pulmonary disease and systemic inflammation: a systematic review and a meta-analysis. Thorax 2004; 59: Kishimoto T. The biology of interleukin-6. Blood 1989; 74: Bucchioni E, Kharitonov SA, Allegra L, et al. High levels of interleukin-6 in the exhaled breath condensate of patients with COPD. Respir Med 2003; 97: Carpagnano GE, Resta O, Foschino-Barbaro MP, et al. Exhaled Interleukine-6 and 8-isoprostane in chronic obstructive pulmonary disease: effect of carbocysteine lysine salt monohydrate (SCMC-Lys). Eur J Pharmacol 2004; 505: Rusznak C, Mills PR, Devalia JL, et al. Effect of cigarette smoke on the permeability and IL-1 and sicam-1 release from cultured human bronchial epithelial cells of neversmokers, smokers, and patients with chronic obstructive pulmonary disease. Am J Respir Cell Mol Biol 23: Di Stefano A, Maestrelli P, Roggeri A, et al. Upregulation of adhesion molecules in the bronchial mucosa of subjects with chronic obstructive bronchitis. Am J Respir Crit Care Med 1994; 149: Beeh KM, Beier J, Kornmann O, et al. Long-term repeatability of induced sputum cells and inflammatory markers in stable, moderately severe COPD. Chest 2003; 123: Ferroni P, Basili S, Martini F, et al. Soluble P-selectin as a marker of platelet hyperactivity in patients with chronic obstructive pulmonary disease. J Investig Med 2000; 48: Schonbeck U, Libby P. CD40 signaling and plaque instability. Circ Res 2001; 89: Schindler C, Ackermann-Liebrich U, Leuenberger P, et al. Associations between lung function and estimated average exposure to NO 2 in eight areas of Switzerland: the SAPALDIA Team; Swiss Study of Air Pollution and Lung Diseases in Adults. Epidemiology 1998; 9: Traves SL, Culpitt SV, Russell RE, et al. Increased levels of the chemokines GRO and MCP-1 in sputum samples from patients with COPD. Thorax 2002; 57: Fiorini G, Crespi S, Rinaldi M, et al. Serum ECP and MPO are increased during exacerbations of chronic bronchitis with airway obstruction. Biomed Pharmacother 2000; 54: CHEST / 133 / 1/ JANUARY,