Circulating alpha1-antitrypsin in the general population: Determinants and association with lung function

1465-9921-9-35 RRJ Research Circulating alpha1-antitrypsin in the general population: Determinants and association with lung function Senn Oliver oliver.senn@usz.ch Russi W Erich erich.russi@usz.ch Schindler Christian christian.schindler@unibas.ch Imboden Medea imboden@medgen.unizh.ch von Eckardstein Arnold arnold.voneckardstein@usz.ch Brändli Otto otto.braendli@zhw.ch Zemp Elisabeth Elisabeth.Zemp@unibas.ch Ackermann-Liebrich Ursula ursula.ackermann-liebrich@unibas.ch Berger Wolfgang berger@medgen.unizh.ch Rochat Thierry thierry.rochat@hcuge.ch Luisetti Maurizio m.luisetti@smatteo.pv.it Probst-Hensch M Nicole nicole.probst@usz.ch the SAPALDIA Team nicole.probst@usz.ch

Molecular Epidemiology/Cancer Registry, Institutes of Social and Preventive Medicine & Clinical Pathology, University of Zurich, Zürich, Switzerland

Pulmonary Division, University Hospital of Zurich, Zürich, Switzerland

Institute of Social and Preventive Medicine, University of Basel, Basel, Switzerland

Institute of Clinical Chemistry, University Hospital of Zurich, Zürich, Switzerland

Zürcher Höhenklinik Wald, Wald, Switzerland

Division of Medical Molecular Genetics and Gene Diagnostics, Institute of Medical Genetics, University of Zurich, Schwerzenbach, Switzerland

Division of Pulmonary Medicine, University Hospital of Geneva, Geneva, Switzerland

Clinica di Malattie dell'Apparato Respiratorio, Laboratorio di Biochimica e Genetica, IRCCS Policlinico San Matteo, Università di Pavia, Pavia, Italy

Respiratory Research 1465-9921 2008 9 1 35 http://respiratory-research.com/content/9/1/35 18439253 10.1186/1465-9921-9-35 15 2 2008 25 4 2008 25 4 2008 2008 Senn et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

Severe alpha1-antitrypsin (AAT) deficiency associated with low AAT blood concentrations is an established genetic COPD risk factor. Less is known about the respiratory health impact of variation in AAT serum concentrations in the general population. We cross-sectionally investigated correlates of circulating AAT concentrations and its association with FEV1.

Methods

In 5187 adults (2669 females) with high-sensitive c-reactive protein (CRP) levels ≤ 10 mg/l from the population-based Swiss SAPALDIA cohort, blood was collected at the time of follow-up examination for measuring serum AAT and CRP.

Results

Female gender, hormone intake, systolic blood pressure, age in men and in postmenopausal women, as well as active and passive smoking were positively, whereas alcohol intake and BMI inversely correlated with serum AAT levels, independent of CRP adjustment. We observed an inverse association of AAT with FEV1 in the total study population (p < 0.001), that disappeared after adjustment for CRP (p = 0.28). In addition, the AAT and FEV1 association was modified by gender, menopausal status in women, and smoking.

Conclusion

The results of this population-based study reflect a complex interrelationship between tobacco exposure, gender related factors, circulating AAT, systemic inflammatory status and lung function.

Introduction

Alpha1-antitrypsin (AAT), a polymorphic inflammation-sensitive plasma protein with antiprotease activity, protects the lung tissue from destruction by neutrophil elastase. Severe and intermediate AAT deficiency is one of the most common inherited diseases on a global scale 1. It increases the risk of chronic obstructive pulmonary diseases (COPD) in active smokers 23. The two most common deficiency alleles are the S- and Z-allele, but at least 30 rare, additional alleles exist that are associated with reduced or absent plasma AAT levels 4.

International AAT deficiency registries have advanced the epidemiologic understanding of genetically determined AAT deficiency. Much less is known about the respiratory health impact of variation in AAT serum concentrations as observed in the general population. According to a study in healthy blood donors only 26% of the variance in circulating AAT was explained by known AAT gene variants 5. In a population-based sample of school children pulmonary function was positively associated with C-reactive protein (CRP)-adjusted AAT levels, independent of PI phenotypes 6. In contrast, an inverse association between inflammation-sensitive proteins, including non-CRP adjusted serum AAT, and pulmonary function has been described in population-based samples of adults 78. These reported opposite directions of the lung function/AAT association may in part be due to differences in adjustment for CRP between the studies. They also agree with the dual role of AAT as both, an antiprotease and an inflammatory marker. Low-grade systemic inflammation is increasingly being recognized as a COPD risk factor 9.

In the population-based Swiss Cohort Study of Air Pollution and Lung Disease in Adults (SAPALDIA) we investigated the potential value of circulating AAT as a biomarker for susceptibility to respiratory disease in the general population. First, we identified environmental and lifestyle factors associated with AAT concentrations in the blood. Second, we investigated the cross-sectional association between circulating AAT concentrations and forced expiratory volume in one second (FEV1).

Methods

Study population

The SAPALDIA cohort has been described 1011. At baseline in 1991 the SAPALDIA participants were aged 18–60 years. They are predominantly of European-Caucasian ethnicity and Swiss nationality. They were randomly selected from eight population registries representing the three major Swiss language regions and including both, urban and rural areas. Health examinations at baseline and after 11-years included an interview about respiratory health, occupational and lifestyle exposures as well as spirometry. Participation rate at baseline was 59.3%. The current, cross-sectional investigation of AAT is restricted to follow-up data collected in 2002/2003 when the blood bank was established (Figure 1 and Table 1 for more detail). Ethical approval for the study was given by the central ethics committee of the Swiss Academy of Medical Sciences and the Cantonal Ethics Committees for each of the eight examination areas.

Figure 1

Participation in SAPALDIA baseline and follow-up examination and in the current cross-sectional study

Participation in SAPALDIA baseline and follow-up examination and in the current cross-sectional study. The current, cross-sectional investigation of AAT is restricted to data collected in 2002/2003 when the blood bank was established. The analysis of factors correlated with serum AAT levels (Tables 2 and 7) is restricted to 5187 follow-up participants (*), whereas the association between serum AAT and FEV1 was further restricted to 4449 follow-up participants with valid lung function measurements (**) (Tables 4, 5 and 6).

Table 1

Characteristics at the baseline examination of participants in different parts of the study

SAPALDIA baseline participants

SAPALDIA participants in study on AAT determinants

SAPALDIA participants in study on AAT and lung function

9651

5187

4449

Female, %

50.7

51.5

48.6

Age (mean, SD) [years]

41.7 (11.7)

40.9 (11.5)

41.1 (11.4)

BMI (mean, SD) [kg/m²]

24.0 (3.9)

23.6 (3.5)

23.7 (3.5)

Smoking Status, %

- never

43.9

49.8

47.8

- former

22.6

20.6

22.0

- current

33.5

29.6

30.2

FEV1 (mean, SD) [ml/s]

3.53 (0.85)

3.59 (0.83)

Measurements

Lung function

Lung function was assessed using spirometry (Sensormedics model 2200, Yorba Linda, USA) according to ATS criteria The present analysis focused on FEV1 expressed as % predicted, derived from SAPALDIA-specific prediction equation 12.

Tobacco exposure, alcohol intake and female hormone-related variables

Information about active and passive smoking was collected by an extended version of the European Community Respiratory Health Survey (ECRHS) questionnaire 13. Smoking was categorized as current (smoking within month before interview), former, and never smoking at follow-up. Never and former smokers were classified as environmental tobacco smoke (ETS) exposed based on an affirmative answer to a question about regular exposure to ETS in the 12 months before the interview. Smoking intensity was calculated as total pack years smoked ([years of smoking * mean cigarettes/day]/20). Current smokers were a priori divided into two subgroups: <15 cigarettes/day vs. ≥ 15 cigarettes/day. Alcohol consumption was a priori classified as daily versus not daily drinking (≥1 vs. <1 alcoholic drink per day).

Women were categorized as pre- and postmenopausal. Women whose menses had become less regular within 12 month before the interview were defined as perimenopausal. Current exogenous hormonal intake was assessed by the questions "have you ever taken oral contraceptives (OC) or hormone replacement therapy (HRT)?", and "were you still taking OC or HRT during the last month before the interview?"

Physical examination

Mean values of 2 automatic measurements of systemic blood pressure (BP) (OMRON 705 CP, Tokyo, Japan) were used and pulse pressure (PP) – a measure of artery stiffness – was calculated as the difference between systolic and diastolic BP 14. Weight and height were measured by electronic (TERRAILLON, Bradford, MA, USA), and telescopic scales (SECA, Hamburg, Germany), respectively for the assessment of body mass index (kg/m²; BMI).

Serum analysis

AAT (g/l) and high-sensitivity c-reactive protein (CRP) (mg/l) concentrations were determined by latex-enhanced immunoturbidimetric assays (Roche Diagnostics, Mannheim, Germany on a Roche COBAS INTEGRA analyzer, Rotkreuz, Switzerland) with interassay coefficients of variation below 5%. Lower detection rates for the AAT and CRP assays were 0.21 g/l, and 0.1 mg/l, respectively.

Statistical analysis and sample size

All statistical analyses were restricted to subjects without signs of an acute infection (CRP ≤ 10 mg/l).

First, we identified the independent association of gender, age, BMI, smoking status, packyears, ETS exposure, and alcohol consumption with AAT concentrations among 5187 participants. Analysis of covariance with and without CRP adjustment was performed to estimate adjusted mean (SE) AAT serum concentrations (g/l).

Second, we investigated the association between circulating AAT and FEV1% predicted. Given the correlations of CRP with both, AAT (Table 2) and FEV1 (Table 3), we investigated the association with and without CPR-adjustment. The regression models included sex, study area, systolic blood pressure, body mass index (BMI), smoking status (never, former, current), alcohol consumption, ETS (never and former smokers), pack years smoked (current smokers), time since quitting (former smokers), cigarettes smoked/day (current smokers), menopausal status and hormone intake (women) as covariates. Analysis of covariance was performed to estimate adjusted mean FEV1% predicted values (standard errors) for AAT quintile classes. To evaluate dose-response relationships, predictor variables were categorized into suitable quantile classes, and regression models without the respective predictor variables were computed. A Cuzick's trend test 15 was then used to test whether regression residuals showed a monotonous association with increasing levels of the predictor in question. Effect modification was assessed by including interaction terms between AAT as a continuous variable and the potential effect modifier of interest in the models. Interactions with a p-value < 0.05 were considered statistically significant 16. In the absence of information on SERPINA1 genotype variants and to avoid unrecognized confounding or effect modification by unrecognized severe and intermediate AAT deficiency genotypes, the main multiple linear regressions were restricted to participants exhibiting normal serum AAT concentrations (≥ 0.9 g/l; indicating the normal cut-off value for AAT serum concentrations according to the consensus reference data [CRM470/RPPHS] 17). The main FEV1 analyses therefore included 4297 subjects (never smokers = 2028, former smokers = 1336, current smokers = 933) with complete data on lung function and covariates. Nevertheless and despite limited sample size, predicted means for the small category of subjects with serum AAT < 0.90 g/l (Q0: n = 152) are also presented in Tables 4, 5 and 6. STATA software, release 8.2 (StataCorp, TX, USA) was used for all statistical analyses.

Table 2

Adjusted* mean AAT serum levels by age, gender, anthropometrics, blood pressure, crp, and lifestyle variables (n = 5187)

Number of subjects

AAT (SE) (g/L), no CRP adjustment

AAT (SE) (g/L), CRP adjustment

Gender

Men

2518

1.203 (0.004)

1.215 (0.004)

Women

2669

1.305 (0.004)

1.293 (0.004)

p < 0.001

Age

Quartiles (years):

Men

<43

630

1.194 (0.007)

1.195 (0.007)

<53

629

1.200 (0.007)

1.205 (0.006)

<61

630

1.208 (0.007)

1.208 (0.006)

<73

629

1.239 (0.007)

1.233 (0.007)

p trend < 0.001

p trend = 0.001

Premenopausal women^¶,†

<38

202

1.314 (0.013)

1.312 (0.012)

<43

201

1.284 (0.013)

1.284 (0.012)

<50

202

1.278 (0.013)

1.281 (0.012)

<55

201

1.227 (0.013)

1.227 (0.012)

p trend < 0.001

Postmenopausal women^¶,†

<55

212

1.221 (0.014)

1.221 (0.013)

<61

212

1.255 (0.013)

1.261 (0.012)

<67

212

1.281 (0.013)

1.276 (0.013)

<73

212

1.275 (0.013)

1.273 (0.013)

p trend 0.05

p trend 0.07

BMI

Quartiles (kg/m²):

<22.79

1338

1.284 (0.005)

1.306 (0.005)

<25.38

1349

1.250 (0.005)

1.257 (0.005)

<28.36

1287

1.241 (0.005)

1.237 (0.005)

≥28.36

1213

1.247 (0.005)

1.218 (0.005)

p trend = 0.003

p trend < 0.001

Systolic blood pressure

Quartiles (mmHg):

<114

1342

1.240 (0.006)

1.244 (0.005)

<126

1279

1.252 (0.005)

1.254 (0.005)

<139

1304

1.260 (0.005)

1.259 (0.005)

≥139

1262

1.272 (0.006)

1.266 (0.005)

p trend < 0.001

p trend 0.002

Tobacco exposure‡

Never smokers/ETS-

2121

1.236 (0.005)

1.238 (0.004)

Never smokers/ETS+

388

1.260 (0.010)

1.269 (0.009)

Former smokers/ETS-

1188

1.238 (0.006)

1.240 (0.005)

Former smokers/ETS+

312

1.249 (0.011)

1.249 (0.010)

Current smokers, <15 cig/d)

515

1.284 (0.008)

1.285 (0.008)

Current smokers, ≥15 cig/d)

662

1.328 (0.009)

1.314 (0.008)

p trend < 0.0001

Alcohol intake

<1 drink per day

4069

1.261 (0.003)

1.262 (0.003)

≥1 drink per day

1118

1.235 (0.006)

1.234 (0.007)

p < 0.001

CRP

Quartiles (mg/l):

<0.6

1506

1.182

<1.1

1202

1.227

<2.2

1232

1.268

≥ 2.2

1247

1.360

P trend < 0.00001

*adjusted for age, gender, smoking status, alcohol intake, BMI, systolic blood pressure and study area ^¶women taking oral contraceptives (n = 166), hormone replacement (n = 524), with perimenopausal status (n = 82), and unclear/missing hormonal status (n = 243) were excluded; ^†p_{(interaction)}pre-versus postmenopausal women <0.001; ‡ n = 5186, due to 1 former smoker with missing ETS exposure; ETS = environmental tobacco exposure

Table 3

Association (z-value; p trend) of FEV1 %predicted with CRP quartile classes^§

z-value#

p_trend(across crp quartiles)

Non-smoking men

FEV1 (% predicted)*

-3.91

<0.000

FEV1 (% predicted)**

-4.53

<0.000

Non-smoking women ¶

FEV1 (% predicted)*

-2.32

0.02

FEV1 (% predicted)**

-2.15

0.03

Smoking men

FEV1 (% predicted)*

-2.44

0.015

FEV1 (% predicted)**

-4.82

<0.000

Smoking women ¶

FEV1 (% predicted)*

-2.44

0.015

FEV1 (% predicted)**

-3.14

0.002

*controlled for BMI, systolic blood pressure, ETS (in non smokers), packyears (in smokers), alcohol intake, study area; ** crude associations; ¶additionally adjusted for menopausal status, and intake of female hormones; subjects with crp > 10 mg/l were excluded; §CRP quartiles (Q) (mg/l): Q1 < 0.6, Q2 < 1.1, Q3 < 2.2, Q4 > 2.2; # Regression models without CRP quartiles as predictor variable were computed first. A Cuzick's trend test was then used to test whether regression residuals showed a monotonous association with CRP quartiles. Under the null hypothesis of no trend, the z-value of Cuzick's trend test statistic approximately follows a standard normal distribution, enabling the computation of approximate p-values; a negative sign indicates an inverse association between CRP and FEV1%predicted [15].

Table 4

Mean (SE) FEV1 %predicted in relation to AAT quintile classes †

Quintile classes of AAT (g/l)

Q0 (<0.9)

Q1 (≥0.9–<1.13)

Q2 (≥1.13, <1.22)

Q3 (≥1.22, <1.31)

Q4 (≥1.31, <1.41)

Q5 (≥1.41)

p_trendfrom model not including Q0

All (n) ¶

152

952

832

932

764

817

FEV1 (% predicted) CRP adj.

99.8 (1.2)

98.0 (0.5)

98.3 (0.5)

98.1 (0.5)

97.6 (0.5)

96.6 (0.5)

0.28

FEV1 (% predicted), no CRP adj. *

100.4 (1.2)

98.7 (0.5)

98.2 (0.5)

97.4 (0.5)

95.5 (0.5)

<0.001

Men (n)‡

627

511

460

351

256

FEV1 (% predicted) CRP adj.

99.0 (1.6)

96.7 (0.6)

97.3 (0.6)

97.2 (0.6)

95.8 (0.8)

93.8 (1.0)

0.08

FEV1 (% predicted), no CRP adj. *

99.3 (1.6)

97.3 (0.6)

97.6 (0.6)

97.1 (0.7)

95.4 (0.8)

92.6 (0.9)

<0.001

Women (n) ¶, ‡

325

321

472

413

561

FEV1 (% predicted) CRP adj.

100.6 (1.6)

99.4 (0.8)

99.2(0.8)

99.3(0.7)

99.2 (0.7)

98.8 (0.6)

0.76

FEV1 (% predicted), no CRP adj. *

101.4 (1.7)

99.9 (0.8)

99.7(0.8)

99.5(0.7)

99.3 (0.7)

97.8 (0.6)

0.19

Never/Former Smokers (n) ¶, ‡

130

811

721

754

541

537

FEV1 (% predicted), CRP adj.

100.4 (1.3)

99.1 (0.5)

99.0 (0.5)

99.1 (0.5)

98.4 (0.6)

96.9 (0.7)

0.10

FEV1 (% predicted), no CRP adj.*

100.9 (1.3)

99.6 (0.5)

99.3 (0.5)

99.1 (0.5)

98.2 (0.6)

95.9 (0.7)

<0.001

Current Smokers (n) ¶, ‡

141

111

178

223

280

FEV1 (% predicted), CRP adj.

96.9 (2.9)

93.6 (1.2)

95.0 (1.3)

93.8 (1.0)

94.5 (0.9)

95.6 (0.9)

0.16

FEV1 (% predicted), no CRP adj.*

97.2 (2.9)

94.3 (1.0)

95.7 (1.3)

94.3 (1.0)

94.7 (0.9)

94.5 (0.9)

0.91

Currents Smokers ≥15 cigs/d (n) ¶,‡

103

122

182

FEV1 (% predicted), CRP adj.

96.5 (4.3)

88.8 (1.9)

93.6 (2.1)

91.2 (1.4)

93.6 (1.3)

95.3 (1.1)

0.01

FEV1 (% predicted), no CRP adj.*

96.7 (4.3)

89.7 (1.9)

94.4 (1.1)

91.2 (1.4)

93.9 (1.3)

94.4 (1.1)

0.06

Currents Smokers <15 cigs/d (n) ¶,‡

101

FEV1 (% predicted), CRP adj.

99.6 (3.9)

98.3 (1.5)

96.9 (1.7)

96.6 (1.5)

95.8 (1.3)

94.9 (1.5)

0.21

FEV1 (% predicted), no CRP adj.*

99.9 (3.9)

98.9 (1.5)

97.6 (1.4)

97.2 (1.5)

95.9 (1.3)

93.5 (1.4)

0.02

† adjusted for gender, smoking status (never, former, current), years since quitting in former smokers, cigarettes/day in current smokers, passive smoking in never smokers, BMI, systolic blood pressure, alcohol consumption, study area, menopausal status in women, intake of female hormones in women, CRP, CRP²(models with CRP adjustment); * adjusted for all covariates but CRP; ¶women in perimenopausal status (n = 35) and missing/unclear status and hormonal intake (n = 120) were excluded from the analysis; ‡ P-values for interactions: In All: AAT*gender: p = 0.02(CRP adj); p = 0.007(no CRP adj.); In women: AAT*menopausal status: 0.01 (CRP adj.); 0.008 no CRP adj.); AAT*intake female hormones: p > 0.80, irrespective of menopausal status; AAT*smoking (heavy current smoking vs. all others): all men: p = 0.02 (CRP adj.); p = 0.03 (no CRP adj.); all women: p = 0.11 (CRP adj.); p = 0.11 (no CRP adj.); postmenopausal women: p = 0.08 (CRP adj.); p = 0.07 (no CRP adj.); premenopausal women: p = 0.59 (CRP adj.); p = 0.58 (no CRP adj.)

Table 5

Mean (SE) FEV1 %predicted in relation to AAT quintile classes in non-current smokers†

Quintile classes of AAT (g/l)

Q0 (<0.9)

Q1 (≥0.9–<1.13)

Q2 (≥1.13, <1.22)

Q3 (≥1.22, <1.31)

Q4 (≥1.31, <1.41)

Q5 (≥1.41)

p_trendfrom model not including Q0

Men (n)

522

439

351

226

142

FEV1 (% predicted), CRP adj.

99.5 (1.8)

97.6 (0.6)

97.9 (0.7)

98.2 (0.8)

97.0 (1.0)

92.9 (1.3)

0.05

FEV1 (% predicted), no CRP adj.*

99.8 (1.8)

98.1 (0.6)

98.0 (0.8)

96.5 (0.9)

91.9 (1.2)

<0.001

Women (n) ¶

289

282

403

315

395

FEV1 (% predicted) CRP adj.

101.0 (1.8)

100.4 (0.9)

99.8 (0.9)

100.3 (0.7)

99.9 (0.8)

99.3 (0.8)

0.85

FEV1 (% predicted), no CRP adj. *

101.8 (1.8)

100.9 (0.9)

100.3 (0.9)

100.4 (0.7)

100.0 (0.8)

98.4 (0.7)

0.13

Premenopausal women (n) ¶

100

109

151

131

187

FEV1 (% predicted), CRP adj.

98.7 (2.4)

98.3 (1.3)

97.2 (1.2)

99.6 (1.0)

101.5 (1.1)

99.7 (1.0)

0.02

FEV1 (% predicted), no CRP adj.*

99.1 (2.4)

98.8 (1.3)

97.1 (1.0)

99.6 (1.0)

101.7 (1.1)

99.1 (1.0)

0.11

Postmenopausal women (n) ¶

189

173

252

184

208

FEV1 (% predicted), CRP adj.

102.5 (2.5)

101.5 (1.1)

101.5 (1.2)

100.8 (1.0)

98.8 (1.1)

98.7 (1.1)

0.04

FEV1 (% predicted), no CRP adj.*

103.4 (2.5)

102.1 (1.1)

101.7 (1.1)

101.0 (1.0)

98.7 (1.1)

97.7 (1.1)

<0.001

† adjusted for BMI, passive smoking, years since quitting, systolic blood pressure, alcohol consumption, study area, CRP, CRP²(models with CRP); * adjusted for all covariates but CRP; ¶additionally adjusted for menopausal status, intake of female hormones; women in perimenopausal status (n = 35) and missing/unclear status and hormonal intake (n = 120) were excluded;

Table 6

Mean (SE) FEV1 %predicted in relation to AAT quintile classes in current smokers †

Quintile classes of AAT (g/l)

Q0 (≤ 0.9)

Q1 (≥0.9–<1.13)

Q2 (≥1.13, <1.22)

Q3 (≥1.22, <1.31)

Q4 (≥1.31, <1.41)

Q5 (≥1.41)

p_trendfrom model not including Q0

Men (n)

105

109

125

114

FEV1 (% predicted), CRP adj.†

95.1 (3.8)

92.8 1.4)

93.8 (1.7)

93.4 (1.3)

93.2 (1.2)

95.4 (1.4)

0.30

FEV1 (% predicted), no CRP adj.*

95.6 (3.8)

93.3 (1.4)

94.1 (1.7)

93.8 (1.3)

93.3 (1.2)

94.3 (1.3)

0.71

Men ≥15 cigs/day (n)

FEV1 (% predicted), CRP adj.†

95.3 (5.7)

89.3 (2.1)

93.3 (2.5)

91.2 (1.7)

93.7 (1.6)

95.5 (1.6)

0.04

FEV1 (% predicted), no CRP adj.*

89.7 (2.1)

93.3 (2.5)

91.4 (1.7)

93.8 (1.7)

94.9 (1.6)

0.07

Men <15 cigs/day (n)

FEV1 (% predicted), CRP adj.†

96.5 (5.0)

96.0 (1.8)

94.3 (2.2)

96.6 (2.2)

92.99 (1.9)

94.1 (2.8)

0.38

FEV1 (% predicted), no CRP adj.*

97.3 (2.2)

96.7 (1.7)

95.2 (2.2)

97.0 (2.2)

92.5 (1.9)

91.1 (2.6)

0.05

Women (n) ¶

166

FEV1 (% predicted), CRP adj.†

98.6 (4.5)

94.6 (2.3)

96.8 (2.2)

94.1 (1.6)

95.9 (1.3)

96.7 (1.1)

0.24

FEV1 (% predicted), no CRP adj.*

98.6 (4.5)

95.5 (2.3)

92.2 (2.2)

94.9 (1.6)

96.4 (1.4)

95.5 (1.1)

0.99

Premenopausal (n)

FEV1 (% predicted), CRP adj.†

98.8 (6.2)

96.1 (2.7)

95.4 (2.7)

97.0 (2.2)

100.2 (2.0)

94.6 (1.4)

0.71

FEV1 (% predicted), no CRP adj.*

98.8 (6.2)

96.7 (2.7)

95.3 (2.2)

99.0 (1.9)

101.0 (1.9)

93.6 (1.4)

0.64

Postmenopausal (n)

FEV1 (% predicted), CRP adj.†

98.3 (6.4)

91.7 (3.7)

97.7 (3.4)

92.4 (2.3)

92.3 (1.8)

99.2 (1.6)

0.15

FEV1 (% predicted), no CRP adj.*

97.1 (6.6)

93.7 (3.7)

99.6 (3.4)

93.5 (2.4)

92.7 (1.9)

97.7 (1.6)

0.69

Postmenopausal, ≥ 15 cigs/day (n))¶

FEV1 (% predicted), CRP adj.†

94.6 (8.3)

84.7 (5.5)

90.0 (6.0)

87.9 (3.6)

90.0 (2.7)

96.3 (2.2)

0.05

FEV1 (% predicted), no CRP adj.*

93.9 (8.5)

87.4 (2.2)

92.7 (6.2)

88.1 (3.7)

90.4 (2.9)

95.1 (2.2)

0.22

Postmenopausal, <15 cigs/day (n)¶

FEV1 (% predicted), CRP adj.†

105.4 (12.3)

100.2 (4.6)

103.6 (3.8)

98.4 (3.0)

96.2 (2.3)

102.3 (2.4)

0.85

FEV1 (% predicted), no CRP adj.*

108.8 (12.6)

100.2 (4.7)

103.5 (3.8)

99.1 (3.0)

96.7 (2.3)

101.3 (2.2)

0.93