Abstract
Background
Systemic inflammation is associated with impaired lung function in healthy adults as well as in patients with lung disease. The mechanism for this association is unknown and it is unclear if systemic inflammation leads to impaired lung function or if poor lung function leads to inflammation. We explored the temporal associations between blood C-reactive protein (CRP), fibrinogen, and white blood cells, and lung function in young adults.
Methods
Spirometry, plethysmography, and diffusion capacity were measured in a population-based cohort at ages 32 and 38 years. High-sensitivity CRP, fibrinogen, and white blood cells were measured at the same ages.
Results
Higher levels of CRP and, to a lesser extent, fibrinogen were associated with lower lung volumes in cross-sectional analyses at both ages 32 and 38 years. Higher CRP and fibrinogen at age 32 were associated with higher FEV1 and FEV1/FVC at age 38, but not other measures of lung function. Lower lung volumes (total lung capacity, functional residual capacity, and residual volume) but not airflow obstruction (FEV1/FVC) at age 32 were associated with higher CRP at age 38. Associations between age 32 lung function and fibrinogen at follow-up were weaker, but consistent. There were no longitudinal associations between white blood cells and lung function.
Conclusions
We found no evidence that systemic inflammation causes a decline in lung function. However, lower lung volumes were associated with higher CRP and fibrinogen at follow-up indicating that pulmonary restriction may be a risk factor for systemic inflammation. The mechanism for this association remains unclear.
Keywords: Cohort studies, C-reactive protein, Inflammation, Pulmonary function
1. Introduction
There is a poorly understood association between systemic inflammation and reduced lung volumes [1–6]. This association is not only present among patients with chronic lung disease, but is also found in apparently healthy young adults. The mechanism(s) behind this association are unknown but may be important for a number of reasons. It is possible that chronic systemic inflammation from a non-respiratory cause leads to an accelerated decline in lung function. Conversely, unrecognised lung disease may result in systemic inflammation with deleterious effects on other aspects of health. Finally, confounding factors, such as obesity, may impact on both lung function and systemic inflammation [7,8]. Since systemic inflammation is implicated in the pathogenesis of cardiovascular diseases, the association may also help to explanation the association between low lung function and cardiovascular mortality [9].
Most studies of lung function and systemic inflammation have been cross-sectional and it remains uncertain which comes first. Two studies found that higher levels of C-reactive protein (CRP) and fibrinogen in young adulthood were associated with a subsequent decline in lung volumes [5,10], but this was not found in other longitudinal studies [2,6,11]. One study found that high CRP levels were associated with subsequent lung function decline only in men [12]. Another suggested that the association between lung function and inflammation may be confounded by adiposity [13]. A recent study confirmed cross-sectional associations between CRP and spirometric lung function, and also found associations between changes in CRP and changes in lung function over 13 years of follow-up. However, baseline CRP levels did not predict lung function decline, nor did baseline lung function predict changes in CRP leaving the temporal sequence between CRP and lung function undetermined [14].
A large Mendelian randomisation study found that, although high levels of CRP were associated with COPD, participants with genetically elevated levels of CRP were not more likely to develop COPD or experience a greater decline in FEV1 during follow-up [15]. This suggests that CRP does not directly mediate a decline in lung function, but it remains possible that CRP levels are an indirect marker of another inflammatory process that impacts on the lungs. A smaller (all male) cohort found an association between baseline CRP levels and FEV1 decline, but this was not predicted by CRP genotype leading the authors to speculate that the association between CRP and lung function could be due to reverse causality –the possibility that impaired lung function causes elevation of CRP [16].
Although there has been considerable research interest in systemic inflammation and obstructive airways disease [17], studies of healthy populations suggest that the association between inflammation and lung function is equally strong for the FVC as the FEV1 and appear to indicate lung restriction rather than airflow obstruction [2,11,14]. One study found an association between systemic inflammation and static lung volumes, but not with airflow obstruction [3]. Most studies have reported spirometry only and we are not aware of any follow-up studies using a broad range of lung function tests.
We previously reported an association between blood CRP levels and reduced spirometric lung volumes in young adults [2].We now report a 6-year follow-up of this cohort with pulmonary function tests of spirometry, static lung volumes, airways conductance, and gas transfer. Our primary aim is to assess whether CRP levels predict a subsequent decline in lung function or whether poor lung function predicts higher CRP at follow-up. We further aim to assess whether inflammation is associated with a restrictive or obstructive physiological impairment. We also considered blood fibrinogen [10,13,18] and white blood cell counts [19] as alternative measures of systemic inflammation that have also been associated with lung function.
2. Methods
Study members were born in Dunedin, New Zealand between April 1972 and March 1973 [20,21]. 1037 children participated in the first follow-up assessment at age 3 years, constituting the base sample of the study. This analysis examines the associations between CRP and lung function at ages 32 and 38 years. At each age, over 95% of living Study members were assessed (972/1015 at age 32 and 961/1007 at age 38), although not all consented to both blood and lung function tests. Study members are mostly of New Zealand/European ethnicity with 7.5% identifying as Māori at age 26 years. Few Study members identified with other ethnicities [22]. The Otago Ethics Committee approved the study. Written informed consent was obtained at each assessment.
Information about respiratory health was obtained at each age using questions from the American Thoracic Society and the European Community Respiratory Health Survey questionnaires [23,24]. Current smokers were defined as those who reported having smoked at least one cigarette a day for at least one month during the previous year. Former smokers were those who had smoked in the past but not within the previous year.
Cumulative smoking was calculated as the pack-years of cigarettes smoked up to each age (20 cigarettes/day for 1 year = 1 pack-year).
Height and weight were measured in light clothing without shoes to calculate Body Mass Index (BMI) in kg/m2. Spirometry (FEV1 and FVC), static lung volumes (total lung capacity (TLC), functional residual capacity (FRC), residual volume (RV)), specific airways conductance (sGaw), single-breath diffusing capacity for carbon monoxide (DLCO), and alveolar volume by methane dilution (VA) were measured to European Respiratory Society/American Thoracic Society standards [25–30] using a body plethysmograph (CareFusion, Yorba Linda, CA). Spirometry was repeated after 200mcg salbutamol via large volume spacer. Participants were asked to avoid using their usual inhalers on the day of the test.
CRP was measured using a high-sensitivity immunoturbidimetric assay (Roche Diagnostics, Germany). Fibrinogen was measured using an automated coagulation analyser (Sysmex; Mahberg, Germany). Haemoglobin and white blood cell counts were measured on an automated analyser (Sysmex Corporation, Japan). Exhaled carbon monoxide was measured twice using a Micro CO monitor, (Micromedical, UK) and the mean value was recorded.
2.1. Statistical methods
Previous analyses of this cohort suggest that associations between CRP and lung function may differ between men and women [2]. Interactions with sexwere therefore investigated for all models. Pregnant women were excluded from all analyses. Because corticosteroids may suppress systemic inflammation and also influence lung function, participants using either oral or inhaled corticosteroids were also excluded.
Linear regression was used to examine cross-sectional associations between inflammation and lung function at ages 32 and 38 using all data available at each age. Analyses were stratified by sex and adjusted for height, BMI, and pack-years of smoking up to that age. Analyses of DLCO and DLCO/VA also adjusted for exhaled carbon monoxide and haemoglobin.
For longitudinal analyses, linear regression was first used to assess associations between CRP at age 32 and changes in lung function at follow-up by using age 38 lung function values as the outcome while adjusting for the same measure at age 32. There was no evidence that sex was an effect modifier in these analyses, so analyses used the combined data set including sex as a covariate. Other covariates were height, changes in BMI, and the pack-years of cigarettes smoked between these assessments. Changes in exhaled carbon monoxide and haemoglobin were included when the outcome was DLCO or DLCO/VA. Interactions between sex and each of baseline CRP, change in pack-years, change in BMI, and (for DLCO and DLCO/VA) change in haemoglobin were investigated and retained where statistically significant.
Next, the direction of the association was reversed and a second set of linear regression models used lung function at 32 to predict CRP levels at age 38 adjusting for age 32 CRP values using the same approach and covariates described above.
CRP was log-transformed for all analyses. Lung function values were also log transformed where this improved the normality and/or homoscedasticity of the residuals. In such cases, ratios of geometric means are reported rather than differences in arithmetic means.
The analyses were repeated using blood fibrinogen [10,13,18] and white blood cell counts. Both were log-transformed for all analyses.
Stata 13.1 (College Station TX) was used with statistical significance determined by two-sided p < 0.05. No adjustments were made for multiple comparisons and interpretation of results focuses on patterns rather than isolated instances of statistical significance.
3. Results
Participant characteristics are shown in Table 1.
Table 1.
Participant characteristics at ages 32 and 38.
| Age 32 | Age 38 | |||
|---|---|---|---|---|
| Men | Women | Men | Women | |
| n = 429 | n = 380 | n = 420 | n = 396 | |
| BMIa | 26.0 (1.2) | 25.3 (1.2) | 27.0 (1.2) | 26.1 (1.2) |
| Smoking: | ||||
| Current smoker | 158 (36.8) | 126 (33.2) | 116 (27.6) | 96 (24.2) |
| Former smoker | 53 (12.4) | 76 (20.0) | 86 (20.5) | 112 (28.3) |
| Never smoker | 218 (50.8) | 178 (46.8) | 218 (51.9) | 188 (47.5) |
| Pack-yearsb | 10.2 (10.9) | 8.2 (8.7) | 12.7 (14.6) | 8.7 (11.7) |
| Inflammation | ||||
| CRP mg/La | 1.1 (2.9) | 1.4 (3.3) | 1.1 (2.6) | 1.4 (3.1) |
| <2 | 326 (76.3) | 230 (60.8) | 314 (74.8) | 249 (62.9) |
| 2–<5 | 67 (15.7) | 85 (22.5) | 77 (18.3) | 87 (22.0) |
| 5–<10 | 22 (5.2) | 41 (10.8) | 21 (5.0) | 36 (9.1) |
| ≥10 | 12 (2.8) | 22 (5.8) | 8 (1.9) | 24 (6.1) |
| Fibrinogen (g/L)a | 2.4 (1.2) | 2.6 (1.2) | 2.6 (1.2) | 2.8 (1.2) |
| WBC (×109/L)a | 7.3 (1.3) | 7.5 (1.3) | 7.5 (1.3) | 7.8 (1.2) |
| Lung function (pre-bronchodilator) | ||||
| FEV1 (L) | 4.5 (0.6) | 3.3 (0.5) | 4.2 (0.6) | 3.2 (0.4) |
| FVC (L) | 5.8 (0.8) | 4.2 (0.6) | 5.6 (0.7) | 4.1 (0.5) |
| FEV1/FVC (%) | 77.3 (6.2) | 80.1 (5.8) | 75.8 (6.1) | 77.7 (5.9) |
| sGaw (mL/s/cmH2O/L)a | 0.2 (1.7) | 0.2 (1.6) | 0.2 (1.4) | 0.2 (1.3) |
| TLC (L)a | 7.4 (1.1) | 5.5 (1.1) | 7.5 (1.1) | 5.6 (1.1) |
| FRC (L) | 3.4 (0.8) | 2.8 (0.6) | 3.1 (0.8) | 2.6 (0.6) |
| RV (L)a | 1.6 (1.3) | 1.3 (1.3) | 1.8 (1.2) | 1.5 (1.3) |
| DLCO (mL/min/mmHg)a | 33.7 (1.2) | 23.7 (1.2) | 31.2 (1.2) | 22.4 (1.2) |
| DLCO/VA (mL/min/mmHg/L)a | 4.8 (0.8) | 4.6 (0.7) | 4.7 (0.7) | 4.5 (0.6) |
Geometric mean (geometric SD).
Median (IQR) for ever smokers.
3.1. Cross sectional associations at ages 32 and 38 years
There were several statistically significant interactions between sex and log-CRP levels and the cross-sectional associations are shown separately (Table 2). Higher log-CRP at age 32 was associated with lower spirometric volumes for both sexes, although the association with FVC was not significant among women. There was a statistically significant inverse association with the FEV1/FVC ratio only among women. Results for men were similar at age 38, but none of the associations between log-CRP and spirometry values were significant for women at this age. A similar pattern of findings was observed for post-bronchodilator spirometry (Appendix Table A.1).
Table 2.
Cross-sectional adjusted associations between CRP and lung function.
| Age 32 | Age 38 | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Men | Women | Sex interaction | Men | Women | Sex interaction | |||||||||||||
| n | Est | 95% CI | p | n | Est | 95% CI | p | p | n | Est | 95% CI | p | n | Est | 95% CI | p | p | |
| FEV1 | 422 | −0.073 | −0.123, −0.023 | 0.004 | 378 | −0.056 | −0.092, −0.021 | 0.002 | 0.446 | 420 | −0.059 | −0.117,−0.002 | 0.043 | 395 | −0.016 | −0.053, 0.021 | 0.389 | 0.205 |
| FVC | 422 | −0.104 | −0.163, −0.046 | 0.001 | 378 | −0.040 | −0.083, 0.003 | 0.071 | 0.029 | 420 | −0.098 | −0.166, −0.029 | 0.006 | 395 | −0.000 | −0.045, 0.045 | 0.989 | 0.002 |
| FEV1/FVC | 422 | 0.128 | −0.443, 0.698 | 0.660 | 378 | −0.601 | −1.146, −0.056 | 0.031 | 0.037 | 420 | 0.260 | −0.403, 0.923 | 0.441 | 395 | −0.370 | −0.962, 0.222 | 0.220 | 0.025 |
| sGawa | 414 | 1.004 | 0.955, 1.055 | 0.879 | 370 | 0.987 | 0.947, 1.030 | 0.553 | 0.835 | 418 | 1.017 | 0.981, 1.054 | 0.362 | 392 | 0.979 | 0.953, 1.006 | 0.134 | 0.354 |
| TLCa | 413 | 0.984 | 0.975, 0.993 | 0.001 | 370 | 0.995 | 0.986, 1.004 | 0.270 | 0.073 | 417 | 0.985 | 0.974,0.996 | 0.006 | 393 | 0.999 | 0.989, 1.010 | 0.869 | 0.082 |
| FRCa | 414 | 0.984 | 0.969, 0.998 | 0.030 | 370 | 0.998 | 0.984, 1.013 | 0.798 | 0.004 | 418 | 0.974 | 0.954, 0.994 | 0.013 | 393 | 0.987 | 0.970, 1.005 | 0.163 | 0.027 |
| RVa | 413 | 0.992 | 0.973, 1.011 | 0.427 | 370 | 1.004 | 0.985, 1.024 | 0.672 | 0.221 | 418 | 0.986 | 0.965, 1.007 | 0.192 | 393 | 1.003 | 0.983, 1.024 | 0.767 | 0.347 |
| DLCOa | 421 | 0.982 | 0.970, 0.995 | 0.006 | 369 | 0.995 | 0.982, 1.007 | 0.394 | 0.261 | 413 | 0.976 | 0.962, 0.990 | 0.001 | 383 | 0.991 | 0.978, 1.003 | 0.148 | 0.138 |
| DLCO/VA | 421 | −0.004 | −0.065, 0.057 | 0.896 | 369 | 0.010 | −0.050, 0.069 | 0.751 | 0.997 | 413 | −0.028 | −0.093, 0.036 | 0.383 | 383 | −0.025 | −0.085, 0.036 | 0.421 | 0.916 |
p values < 0.05 are highlighted in bold.
Log-transformed outcome so estimates and CIs are ratios of geometric means rather than differences of arithmetic means. Log-transformed CRP was used as the independent variable of interest. Analyses are adjusted for height, BMI, and pack years smoking. DLCO and DLCO/VA analyses are also adjusted for exhaled carbon monoxide and haemoglobin.
Negative associations were found between log-CRP and TLC and FRC for men at both ages but not for women. Negative associations between log-CRP and DLCO were also found for men but not women, but the ratio of DLCO/VA was not associated with log-CRP in either sex.
The associations between lung function and fibrinogen were weaker and there were few statistically significant findings. There were no significant associations between fibrinogen and lung function in men at age 32. In women at 32, FEV1, FVC, TLC and DLCO were inversely associated with fibrinogen. In men at age 38 FVC and DLCO were inversely associated with fibrinogen. In women at age 38, DLCO was inversely associated with fibrinogen (Appendix Table A.2). White blood cell counts had even weaker associations with lung function: the only statistically significant findings were inverse associations with FEV1 and DLCO at age 32 in women (Appendix Table A.2).
3.2. Longitudinal associations
(i) Does baseline CRP predict lung function at follow up?
Higher baseline log-CRP was associated with greater FEV1 values and FEV1/FVC ratios at age 38. None of the associations between baseline log-CRP and other measures of lung function at age 38 were significant (Table 3).
Table 3.
Associations between baseline values of CRP and changes in lung function measures.
| n | Estimate | 95% CI | p-value | |
|---|---|---|---|---|
| FEV1b | 729 | 0.020 | 0.006, 0.034 | 0.005 |
| FVCb | 729 | 0.005 | −0.012, 0.021 | 0.586 |
| FEV1/FVCb | 729 | 0.340 | 0.130, 0.550 | 0.002 |
| sGawa | 716 | 0.994 | 0.980, 1.007 | 0.353 |
| TLCa,b | 714 | 1.001 | 0.998, 1.004 | 0.358 |
| FRCa,b | 716 | 0.997 | 0.989, 1.006 | 0.581 |
| RVa | 715 | 0.996 | 0.985, 1.006 | 0.384 |
| DLCOa | 698 | 1.002 | 0.997, 1.008 | 0.409 |
| DLCO/VA | 698 | 0.023 | −0.003, 0.049 | 0.088 |
p values < 0.05 are highlighted in bold.
Log-transformed outcome so estimates and CIs are ratios of geometric means rather than differences of arithmetic means. Log-transformed CRP was used as the independent variable of interest Analyses are adjusted for sex, height, changes in BMI and pack years smoking between ages 32 and 38 years. DLCO and DLCO/VA analyses are also adjusted for changes in exhaled carbon monoxide and haemoglobin.
Sex-BMI interaction included in the model.
There was also an association between higher log-CRP values at baseline and higher post-bronchodilator FEV1/FVC ratios at follow-up but other post-bronchodilator spirometry measures were not significantly associated with baseline CRP (Appendix Table A.4).
The longitudinal associations for fibrinogen and white blood cells are shown in Appendix Table A.5. Similar to CRP, higher fibrinogen at age 32 predicted higher FEV1/FVC ratios at age 38. White blood cell counts at age 32 were not significantly associated with lung function changes at age 38 aside from a negative association with RV (Appendix Table A.6).
(ii) Does baseline lung function predict CRP at follow-up?
Higher baseline lung volumes (TLC, FRC, and RV) were associated with lower follow-up log-CRP (Table 4). Gas transfer (DLCO) was also associated with higher log-CRP at follow-up, but not when it was corrected for alveolar volume (DLCO/VA). Neither pre- nor post-bronchodilator spirometry at baseline were associated with changes in CRP at follow up (Table 4 and Appendix Table A.7).
Table 4.
Associations between baseline values of lung function measures and changes in CRP.
| n | Estimate | 95% CI | p-value | |
|---|---|---|---|---|
| FEV1 | 718 | 0.902 | 0.786, 1.035 | 0.142 |
| FVC | 718 | 0.915 | 0.817, 1.026 | 0.127 |
| FEV1/FVC | 718 | 1.000 | 0.989, 1.011 | 0.959 |
| sGaw | 711 | 1.221 | 0.769, 1.937 | 0.397 |
| TLC | 710 | 0.895 | 0.815, 0.983 | 0.021 |
| FRC | 711 | 0.758 | 0.684, 0.840 | <0.001 |
| RV | 710 | 0.793 | 0.654, 0.962 | 0.018 |
| DLCO | 703 | 0.981 | 0.966, 0.996 | 0.012 |
| DLCO/VA | 703 | 0.976 | 0.890, 1.070 | 0.605 |
CRP was log-transformed so estimates and CIs are ratios of geometric means rather than differences of arithmetic means. Analyses are adjusted for sex, height, changes in BMI and pack years smoking between ages 32 and 38 years. DLCO and DLCO/VA analyses are also adjusted for changes in exhaled carbon monoxide and haemoglobin.
p values < 0.05 are highlighted in bold.
The associations between baseline lung function and change in fibrinogen levels are shown in Appendix Table A.8. Higher levels of FRC were associated with lower levels of fibrinogen at follow-up, but there were no other significant associations. There were no associations between baseline lung function and change in white blood cell counts (Appendix Table A.9).
3.3. Analysis among never-smokers
The analyses were repeated on the participants (n = 396 at age 32 and n = 406 at age 38) who had never smoked to exclude the possibility of residual confounding by smoking status. The results for CRP were similar to the full cohort with the following changes. The cross-sectional models revealed attenuated and non-significant associations at age 32 for men with respect to FEV1 and FRC, and for women for FEV1 and FEV1/FVC. For women at age 38, the association with FEV1/FVC was now statistically significant. In the longitudinal models, the association between baseline log-CRP and change in FEV1 was attenuated and not statistically significant, while the inverse association between baseline DLCO and CRP at follow up was unchanged, but not statistically significant (results not shown).
4. Discussion
Cross sectional associations at both age 32 and 38 years in this population-based cohort confirm previous observations of an association between systemic inflammation and lower levels of lung function and extend these observations to include a broad range of lung function tests [1–6,14]. These were more apparent in men than women and were observed across a range of lung volume measurements. Although there were cross-sectional associations between higher CRP levels and lower FEV1/FVC ratios in women, the associations in men were more consistent with a restrictive rather than obstructive pattern. None of the associations with specific airway conductance (sGaw) were significant. Similarly, the association between CRP and lower DLCO in men was not found when corrected for alveolar volume (DLCO/VA) suggesting that the association was primarily due to reduced lung volume rather than impaired gas transfer per se.
From the longitudinal models, we did not find any evidence that systemic inflammation predicted a decline in lung function at follow-up. In fact higher levels of CRP at age 32 were associated with higher FEV1 and FEV1/FVC ratios at age 38. Conversely, lower lung volumes at baseline predicted higher CRP levels at follow-up. These associations were found for plethysmographic lung volumes, but not for spirometry or airflow obstruction indicating that higher levels of systemic inflammation at follow up are predicted by a restrictive rather than obstructive lung function pattern.
Only a few studies have evaluated the temporal sequence between inflammation and reduced lung volumes. Some studies found no association between baseline CRP and decline in FEV1 or FVC levels over 6–13 years follow-up [11,14,31], whereas other studies reported associations between higher levels of CRP at baseline and decline in spirometric lung function over follow-up of between 9 and 27 years [5,10,16,32]. Fewer studies have explored whether lung function at baseline predicts the subsequent change in CRP. We have previously found that a decline in FEV1 between ages 26 and 32 was associated with an increase in CRP, but this analysis was limited by the use of a low-sensitivity CRP assay at baseline [2]. Other studies have found no association between baseline FEV1 and change in CRP [6,14], while a study in young Danish adults found that whereas baseline FEV1 did not predict the change in CRP, the decline in FEV1 predicted CRP at follow-up [5].
We found that associations between lung function and fibrinogen were generally weaker than the associations between lung function and CRP in both the cross-sectional and longitudinal analyses, whereas there were very few significant associations between lung function and white blood cell counts (Appendix Tables). Although, most of the identified associations were in a similar direction to those seen for CRP, this suggests that these measures of inflammation are less clearly associated with lung function. In contrast to our findings, a longitudinal study of elderly adults found that baseline fibrinogen, but not CRP, predicted a decline in the FEV1/FVC ratio [18]. In keeping with our findings, however, both CRP and fibrinogen were associated with higher levels of lung function at follow-up, which was interpreted as due to a survival bias. This would not apply to our younger cohort.
Our findings do not support the hypothesis that low-grade inflammation leads to lung damage. This is consistent with a large Mendelian randomisation study, which found that although high CRP levels were associated with worse lung function overall, those who were genetically predisposed to higher CRP levels did not have worse lung function [15]. Our findings extend these observations by showing that it is unlikely that CRP is a marker of an underlying inflammatory process that is causing lung damage. In fact, we found the opposite tendency – those with higher CRP at baseline tended to have improved spirometry values 6 years later. It seems unlikely that higher levels of CRP directly lead to improved lung function and one possible explanation is that a sub-clinical illness at the time of testing at age 32 could have led to both a higher CRP level and temporarily impaired spirometry.
By contrast, we found that lower lung volumes at baseline predicted higher inflammatory responses six years later. These findings were similar when the analyses were restricted to never smokers. This association could not be explained by the potential confounding influences of changes in body weight or smoking. It suggests that poor lung function leads to increased systemic inflammation, although whether this is a direct influence of the lungs on the inflammatory response or mediated by a third factor is unknown. The finding does, however, offer a potential explanation for the association between reduced lung volumes and cardiac disease, since CRP may be implicated in the pathogenesis of atherosclerosis [33].
There were sex-differences in the cross-sectional associations between poor lung function and systemic inflammation. For women, the associations were weaker than in men and generally not statistically significant. Several studies have found stronger association between CRP and lung function in men [4,5,32]. A possible explanation for this finding is that CRP tends to be higher in women and may fluctuate to a greater extent due to hormonal changes with the menstrual cycle. However, the strength of the association between CRP levels at 32 and 38 years was similar in men (Spearman's rho = 0.49) and women (rho = 0.53) suggesting that CRP levels are equally stable in both sexes. Hence, we cannot easily explain the sex difference and it remains to be seen if this persists with advancing age in future assessments of the cohort. We note, however, that the association between CRP and cardiovascular disease appears to be stronger in men than women, and this is similar to our finding of stronger associations between CRP and lung function in men [34].
Strengths of this study include a broad range of measures of lung function taken six years apart in a population-based cohort with a very high follow-up rate. The fact that all Study members were the same age eliminates confounding by age. We also have information on potential confounding variables. A limitation of the study is that 6 years of follow-up data may not reflect changes in inflammation and lung function over a longer time.
In conclusion, cross-sectional analyses confirm the association between systemic inflammation and reduced lung function. Longitudinal analyses, however, provide no support for the hypothesis that systemic inflammation leads to lung function impairment. Conversely, lower lung volumes were associated with higher levels of inflammation at follow-up.
Supplementary Material
Acknowledgments
We are grateful to the Study members and their friends and families for their continued support. We also thank Dr Phil A Silva, the study founder. The Dunedin Multidisciplinary Health and Development Research Unit is funded by the Health Research Council of New Zealand. The US-National Institute of Aging grant AG032282 provided additional support. We are grateful to professors Terrie Moffit and Avshalom Caspi for sharing data used in this report.
Footnotes
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.rmed.2015.12.007.
Conflict of interest
None of the authors has a conflict of interest with the subject of this manuscript.
Contributors
RJH developed the idea for this study and was involved in data collection, analysis, and preparation of the manuscript. ARG was involved in the planning of the analyses, analysed the data, and helped to prepare the manuscript. MRS and RP collected data, contributed to the analyses, and critically reviewed the manuscript. All authors approved the manuscript prior to submission.
References
- 1.Aronson D, Roterman I, Yigla M, Kerner A, Avizohar O, Sella R, Bartha P, Levy Y, Markiewicz W. Inverse association between pulmonary function and c-reactive protein in apparently healthy subjects. Am. J. Respir. Crit. Care Med. 2006;174:626–632. doi: 10.1164/rccm.200602-243OC. [DOI] [PubMed] [Google Scholar]
- 2.Hancox RJ, Poulton R, Greene JM, Filsell S, McLachlan CR, Rasmussen F, Taylor DR, Williams MJ, Williamson A, Sears MR. Systemic inflammation and lung function in young adults. Thorax. 2007;62:1064–1068. doi: 10.1136/thx.2006.076877. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gläser S, Ittermann T, Koch B, Völzke H, Wallaschofski H, Nauck M, Warnke C, Vogelmeier C, Schulz H, Felix SB, Ewert R, Schäper C. Airflow limitation, lung volumes and systemic inflammation in a general population. Eur. Respir. J. 2012;39:29–37. doi: 10.1183/09031936.00009811. [DOI] [PubMed] [Google Scholar]
- 4.Thorleifsson SJ, Margretardottir OB, Gudmundsson G, Olafsson I, Benediktsdottir B, Janson C, Buist AS, Gislason T. Chronic airflow obstruction and markers of systemic inflammation: results from the BOLD study in Iceland. Respir. Med. 2009;103:1548–1553. doi: 10.1016/j.rmed.2009.04.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Rasmussen F, Mikkelsen D, Hancox RJ, Lambrechtsen J, Nybo M, Hansen HS, Siersted HC. High-sensitive C-reactive protein is associated with reduced lung function in young adults. Eur. Respir. J. 2009;33:382–388. doi: 10.1183/09031936.00040708. [DOI] [PubMed] [Google Scholar]
- 6.Shaaban R, Kony S, Driss F, Leynaert B, Soussan D, Pin I, Neukirch F, Zureik M. Change in C-reactive protein levels and FEV1 decline: a longitudinal population-based study. Respir. Med. 2006;100:2112–2120. doi: 10.1016/j.rmed.2006.03.027. [DOI] [PubMed] [Google Scholar]
- 7.Littleton SW. Impact of obesity on respiratory function. Respirology. 2012;17:43–49. doi: 10.1111/j.1440-1843.2011.02096.x. [DOI] [PubMed] [Google Scholar]
- 8.Williams MJ, Williams SM, Milne BJ, Hancox RJ, Poulton R. Association between C-reactive protein, metabolic cardiovascular risk factors, obesity and oral contraceptive use in young adults. Int. J. Obes. Relat. Metab. Disord. 2004;28:998–1003. doi: 10.1038/sj.ijo.0802713. [DOI] [PubMed] [Google Scholar]
- 9.Agusti Al, Soriano JB. COPD as a systemic disease. COPD J. Chronic Obstr. Pulm. Dis. 2008;5:133–138. doi: 10.1080/15412550801941349. [DOI] [PubMed] [Google Scholar]
- 10.Kalhan R, Tran BT, Colangelo LA, Rosenberg SR, Liu K, Thyagarajan B, Jacobs DR, Jr, Smith LJ. Systemic inflammation in young adults is associated with abnormal lung function in middle age. PLoS One. 2010;5:e11431. doi: 10.1371/journal.pone.0011431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Fogarty AW, Jones S, Britton JR, Lewis SA, McKeever TM. Systemic inflammation and decline in lung function in a general population: a prospective study. Thorax. 2007;62:515–520. doi: 10.1136/thx.2006.066969. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Olafsdottir IS, Gislason T, Thjodleifsson B, Olafsson I, Gislason D, Jogi R, Janson C. Gender differences in the association between C-reactive protein, lung function impairment, and COPD. Int. J. Chron. Obstruct Pulm. Dis. 2007;2:635–642. [PMC free article] [PubMed] [Google Scholar]
- 13.Pearce MS, Hancox RJ, Parker L, Gibson GJ. Lung function and plasma fibrinogen concentrations in the newcastle thousand families birth cohort between age 49 and 51 years. Respirology. 2014;19:53–57. doi: 10.1111/resp.12158. [DOI] [PubMed] [Google Scholar]
- 14.Ahmadi-Abhari S, Kaptoge S, Luben RN, Wareham NJ, Khaw K-T. Longitudinal association of C-reactive protein and lung function over 13 years: the EPIC-Norfolk study. Am. J. Epidemiol. 2014;179:48–56. doi: 10.1093/aje/kwt208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Dahl M, Vestbo J, Zacho J, Lange P, Tybjærg-Hansen A, Nordestgaard BG. C reactive protein and chronic obstructive pulmonary disease: a mendelian randomisation approach. Thorax. 2011;66:197–204. doi: 10.1136/thx.2009.131193. [DOI] [PubMed] [Google Scholar]
- 16.Bolton CE, Schumacher W, Cockcroft JR, Timpson NJ, Smith GD, Gallacher J, Rumley A, Lowe GD, Ebrahim S, Shale DJ, Ben-Shlomo Y. The CRP genotype, serum levels and lung function in men: the caerphilly prospective study. Clin. Sci. 2011;120:347–355. doi: 10.1042/CS20100504. [DOI] [PubMed] [Google Scholar]
- 17.Thomsen M, Dahl M, Lange P, Vestbo J, Nordestgaard BG. Inflammatory biomarkers and comorbidities in chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 2012;186:982–988. doi: 10.1164/rccm.201206-1113OC. [DOI] [PubMed] [Google Scholar]
- 18.Jiang R, Burke GL, Enright PL, Newman AB, Margolis HG, Cushman M, Tracy RP, Wang Y, Kronmal RA, Barr RG. Inflammatory markers and longitudinal lung function decline in the elderly. Am. J. Epidemiol. 2008;168:602–610. doi: 10.1093/aje/kwn174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.McKeever T, Saha S, Fogarty AW. The association between systemic inflammatory cellular levels and lung function: a population-based study. PLoS One. 2011;6:e21593. doi: 10.1371/journal.pone.0021593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Silva PA, Stanton WR. From Child to Adult: the Dunedin Multidisciplinary Health and Development Study. Auckland: Oxford University Press; 1996. [Google Scholar]
- 21.Sears MR, Greene JM, Willan AR, Wiecek EM, Taylor DR, Flannery EM, Cowan JO, Herbison GP, Silva PA, Poulton R. A longitudinal, population-based, cohort study of childhood asthma followed to adulthood. N. Engl. J. Med. 2003;349:1414–1422. doi: 10.1056/NEJMoa022363. [DOI] [PubMed] [Google Scholar]
- 22.Hancox RJ, Poulton R, Greene JM, McLachlan CR, Pearce MS, Sears M. Associations between birth weight, early childhood weight gain, and adult lung function. Thorax. 2009;64:228–232. doi: 10.1136/thx.2008.103978. [DOI] [PubMed] [Google Scholar]
- 23.Ferris BG. Epidemiology standardization project (American thoracic society) Am. Rev. Respir. Dis. 1978;118:1–120. [PubMed] [Google Scholar]
- 24.Burney P, Chinn S. Developing a new questionnaire for measuring the prevalence and distribution of asthma. Chest. 1987;91:79S–83S. doi: 10.1378/chest.91.6_supplement.79s. [DOI] [PubMed] [Google Scholar]
- 25.American Thoracic Society. Standardization of spirometry, 1994 update. Am. J. Respir. Crit. Care Med. 1995;152:1107–1136. doi: 10.1164/ajrccm.152.3.7663792. [DOI] [PubMed] [Google Scholar]
- 26.American Thoracic Society. Single-breath carbon monoxide diffusing capacity (transfer factor). Recommendations for a standard technique-1995 update. Am. J. Respir. Crit. Care Med. 1995;152:2185–2198. doi: 10.1164/ajrccm.152.6.8520796. [DOI] [PubMed] [Google Scholar]
- 27.Coates AL, Peslin R, Rodenstein D, Stocks J. Measurement of lung volumes by plethysmography. Eur. Respir. J. 1997;10:1415–1427. doi: 10.1183/09031936.97.10061415. [DOI] [PubMed] [Google Scholar]
- 28.MacIntyre N, Crapo RO, Viegi G, Johnson DC, van der Grinten CPM, Brusasco V, Burgos F, Casaburi R, Coates A, Enright P, Gustafsson P, Hankinson J, Jensen R, McKay R, Miller MR, Navajas D, Pedersen OF, Pellegrino R, Wanger J. Standardisation of the single-breath determination of carbon monoxide uptake in the lung. Eur. Respir. J. 2005;26:720–735. doi: 10.1183/09031936.05.00034905. [DOI] [PubMed] [Google Scholar]
- 29.Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, Crapo R, Enright P, van der Grinten CPM, Gustafsson P, Jensen R, Johnson DC, MacIntyre N, McKay R, Navajas D, Pedersen OF, Pellegrino R, Viegi G, Wanger J. Standardisation of spirometry. Eur. Respir. J. 2005;26:319–338. doi: 10.1183/09031936.05.00034805. [DOI] [PubMed] [Google Scholar]
- 30.Wanger J, Clausen JL, Coates A, Pedersen OF, Brusasco V, Burgos F, Casaburi R, Crapo R, Enright P, van der Grinten CPM, Gustafsson P, Hankinson J, Jensen R, Johnson D, MacIntyre N, McKay R, Miller MR, Navajas D, Pellegrino R, Viegi G. Standardisation of the measurement of lung volumes. Eur. Respir. J. 2005;26:511–522. doi: 10.1183/09031936.05.00035005. [DOI] [PubMed] [Google Scholar]
- 31.Nybo M, Hansen HS, Siersted HC, Rasmussen F. No relationship between lung function and high-sensitive C-reactive protein in adolescence. Clin. Respir. J. 2010;4:230–236. doi: 10.1111/j.1752-699X.2009.00181.x. [DOI] [PubMed] [Google Scholar]
- 32.Ólafsdóttir IS, Gíslason T, Gudnason V, Benediktsdóttir B, Ólafsson Í, Aspelund T, Thjódleifsson B, Janson C. CRP is associated with lung function decline in men but not women: a prospective study. Respir. Med. 2013;107:91–97. doi: 10.1016/j.rmed.2012.09.020. [DOI] [PubMed] [Google Scholar]
- 33.Yousuf O, Mohanty BD, Martin SS, Joshi PH, Blaha MJ, Nasir K, Blumenthal RS, Budoff MJ. High-sensitivity C-reactive protein and cardiovascular disease: a resolute belief or an elusive link? J. Am. Coll. Cardiol. 2013;62:397–408. doi: 10.1016/j.jacc.2013.05.016. [DOI] [PubMed] [Google Scholar]
- 34.Emerging Risk Factors Collaboration. Kaptoge S, Di Angelantonio E, Pennells L, Wood AM, White IR, Gao P, Walker M, Thompson A, Sarwar N, Caslake M, Butterworth AS, Amouyel P, Assmann G, Bakker SJL, Barr ELM, Barrett-Connor E, Benjamin EJ, Björkelund C, Brenner H, Brunner E, Clarke R, Cooper JA, Cremer P, Cushman M, Dagenais GR, D'Agostino RB, Dankner R, Davey-Smith G, Deeg D, Dekker JM, Engström G, Folsom AR, Fowkes FGR, Gallacher J, Gaziano JM, Giampaoli S, Gillum RF, Hofman A, Howard BV, Ingelsson E, Iso H, Jørgensen T, Kiechl S, Kitamura A, Kiyohara Y, Koenig W, Kromhout D, Kuller LH, Lawlor DA, Meade TW, Nissinen A, Nordestgaard BG, Onat A, Panagiotakos DB, Psaty BM, Rodriguezv B, Rosengren A, Salomaa V, Kauhanen J, Salonen JT, Shaffer JA, Shea S, Ford I, Stehouwer CDA, Strandberg TE, Tipping RW, Tosetto A, Wassertheil-Smoller S, Wennberg P, Westendorp RG, Whincup PH, Wilhelmsen L, Woodward M, Lowe GDO, Wareham NJ, Khaw K-T, Sattar N, Packard CJ, Gudnason V, Ridker PM, Pepys MB, Thompson SG, Danesh J. C-reactive protein, fibrinogen, and cardiovascular disease prediction. N. Engl. J. Med. 2012;367:1310–1320. doi: 10.1056/NEJMoa1107477. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.