Change in Lung Function and Morbidity from Chronic Obstructive Pulmonary Disease in 1-Antitrypsin MZ Heterozygotes: A Longitudinal Study of the General Population

Context The clinical importance of 1-antitrypsin heterozygosity is uncertain. Contribution This Danish population-based study followed 9187 participants (451 of whom were heterozygotes) with questionnaires and rep eated pulmonary function testing for up to 18 years. Compared with persons who had the normal (MM) genotype, heterozygotes had a slightly greater rate of decrease in lung function, slightly higher incidence of chronic obstructive pulmonary disease (COPD), and slightly higher occurrence of hospitalization or death from COPD. Implications Because the incidence of 1-antitrypsin heterozygosity in the general population is much higher than that of homozygosity, 1-antitrypsin heterozygosity is a more important public health problem than homozygosity. The Editors Chronic obstructive pulmonary disease (COPD) is one of the most important health problems worldwide (1, 2). More than 90% of COPD cases are caused by smoking, but only a fraction of smokers develop this disease. The variability in COPD expression among smokers could be due to differences in the environment, in genetic predisposition, or both. So far, severe 1-antitrypsin deficiency has been the best described genetic risk factor for COPD (2, 3). When lung tissue is 1-antitrypsin deficient, protection from neutrophil elastase is impaired and elastic tissue is slowly destroyed, ultimately leading to reduced lung function and development of COPD (4). Severe 1-antitrypsin deficiency is almost entirely caused by the presence of the Z alleles in the 1-antitrypsin gene rather than the normal M allele. Relative plasma 1-antitrypsin concentrations are approximately 16% for persons with the ZZ genotype, 83% for persons with the MZ genotype, and 100% for persons with the MM genotype (5). A deteriorating effect of severe 1-antitrypsin deficiency (the ZZ genotype) on lung function has been known for many years (2-4), whereas the role of intermediate deficiency (the MZ genotype) remains uncertain (6, 7). We previously performed a cross-sectional study of the at-large population in Denmark that assessed the MZ genotype and risk for lung disease. We found that MZ heterozygosity was associated with reduced pulmonary function in persons with clinically established COPD but not in persons without the disease (who account for 98% of the population) (8). However, MZ heterozygosity may lead to increased annual loss of lung function and increased COPD risk in the average person during prolonged follow-up. Furthermore, other genetic variability in the 1-antitrypsin gene may influence the association between the MZ genotype and lung function and disease. Although we previously considered the other well-known structural mutation in 1-antitrypsin, the S allele, we had not considered the effect of the known variation in the 1-antitrypsin promotor, the E allele (9). Finally, genetic variation in the cystic fibrosis gene may also influence lung function and disease (10), and we had not considered this in our former work (8). The Epolymorphism is a known (9) but infrequently studied variation of the 1-antitrypsin gene that is situated in the 3 noncoding enhancer binding region. This polymorphism may temper the increase in plasma 1-antitrypsin concentrations that normally occurs during an acute-phase response but does not affect the baseline levels (11). The more severe Z and S polymorphisms, which are located in the coding regions of the gene, cause the protein to self-polymerize in the liver before secretion into the blood, thereby reducing baseline 1-antitrypsin concentrations (4, 12). Using 21-year follow-up data and triple measurements of lung function (obtained in 1976 to 1978, 1981 to 1983, and 1991 to 1994), we now report on the relationship between MZ heterozygosity and pulmonary function and disease in greater detail. We compared 10 different genotype combinations of the Z, S, E, and M alleles with respect to 1-antitrypsin levels in plasma, annual decrease in FEV1, spirometry-defined airway obstruction, and hospitalization or death from COPD. We also examined whether the common F508 mutation in the cystic fibrosis gene influences these associations. For these purposes, we studied a sample of 9187 persons in the Danish general population from the Copenhagen City Heart Study. Methods Participants and Study Protocol We studied participants in the Copenhagen City Heart Study, a prospective epidemiologic study in Denmark initiated in 1976 to 1978 (13, 14). A sample of 19 698 persons at least 20 years of age was randomly selected after stratification of the Copenhagen residents into 5-year age groups. From 1976 to 1978, participants were invited to complete a survey and then undergo a physical examination at Copenhagen University Hospital. All participants were subsequently invited to participate in a second questionnaire and physical examination from 1981 through 1983 and a third survey and examination from 1991 through 1994. Additional persons in the youngest age group were invited to participate after 5 years (n = 500) and 15 years (n = 3000). A total of 14 223 participants (response rate, 74%) attended the first examination, 12 698 (response rate, 70%) attended the second examination, and 10 135 (response rate, 61%) attended the third examination. Of the 10 135 persons who participated in the third survey and examination, 9259 provided blood samples and 9187 underwent genotyping for the M, Z, S, and E alleles and the F508 mutation in the cystic fibrosis gene (10). Details of the procedures for sample selection and the examination and characteristics of the persons who did not respond to the study survey are described elsewhere (8, 13, 14). Less than 1% of the participants were not white, and 99% were of Danish descent. All participants gave informed consent before entering the study. The ethics committee for the cities of Copenhagen and Frederiksberg approved the study. Study Questionnaire Participants filled out a self-administered questionnaire. At the study examination, an investigator and the participant scrutinized this questionnaire to ensure that the responses were accurate. All participants reported whether they were current smokers, ex-smokers, or never-smokers. Lifetime tobacco exposure before study entry was estimated in pack-years by multiplying daily tobacco consumption (in g) by the smoking duration (in years) and then dividing this number by 20 (g/pack). Smoking during follow-up (in g/d) was determined at each examination by self-report of the average amount of tobacco consumed daily. Spirometry At the first and second examinations, FEV1 and forced vital capacity (FVC) were measured by using an electronic spirometer (model N 403, Monaghan, Littleton, Colorado), which was calibrated daily with a 1-L syringe and was calibrated weekly against a water-sealed Godard spirometer. The instrument used at the third examination was a dry-wedge spirometer (Vitalograph, Maids Moreton, Buckinghamshire, United Kingdom) that was calibrated daily with a 1-L syringe. To ensure accurate measurements, three sets of measurements were obtained at each examination, and the values produced on at least two of the measurements obtained at an examination had to differ by less than 5%. We used the highest set of FEV1 and FVC at each of the three examinations in our analyses as absolute values and as percentage of predicted values by using internally derived reference values based on a subsample of healthy never smokers [15]. We considered participants to have airway obstruction if they met each of the following criteria at least once during the study: 1) FEV1 less than 80% of the predicted value and 2) FEV1/FVC less than 0.7 (16). Most participants with airway obstruction had COPD; however, 22% of the participants with airway obstruction reported having asthma. Whether self-reports of asthma were based on a physician's diagnosis or the participant's misconception of COPD could not be evaluated. To calculate annual change in FEV1 (in mL/y), the most recently obtained FEV1 (in mL) was subtracted by the FEV1 value obtained at the first measurement, this difference was multiplied by 365.25, and this product was divided by the number of days between the two FEV1 measurements (in years 1). COPD Study classification of COPD required COPD as the main diagnosis by a physician at discharge or on the death certificate. The information on COPD diagnoses, which were drawn from the Danish National Hospital Discharge Register (data from 1976 to 1997 were available) and the Danish Register of Causes of Death (data from 1992 to 1999 were available), were based on the World Health Organization International Classification of Diseases, 8th or 10th edition (diagnosis codes 4902 and J404, respectively) (17, 18). Genotype Analysis We identified the Z (342GluLys), S (264GluVal), and E (1237GA) (9) polymorphisms in the 1-antitrypsin gene by performing polymerase chain reaction (PCR). We used the 5 ATAAGGCTGTGCTGACCATCGTC 3 (sense) and 5 TTGGGTGGGATTCACCACTTTTC 3 (antisense) primers to identify the Z allele, the 5 TGAGGGGAAACTACAGCACCTCG 3 (sense) and 5 AGGTGTGGGCAGCTTCTTGGTCA 3 (antisense) primers to identify the S allele, and the 5 GTTCCTGAATAGCCCCTGTGGTA 3 (sense) and 5 CGGTATCCATTGATTAGACTGAA 3 (antisense) primers to identify the E allele. The presence of the three polymorphisms (Z, S, and E) destroyed a Taq1 site in the respective PCR products. Fragments of 157 base pair (bp) and 22 bp (normal allele) or 179 bp (Z allele), 100 bp and 21 bp (normal allele) or 121 bp (S allele), and 258 bp and 59 bp (normal allele) or 317 bp (E allele) were separated on a 3% agarose gel. The F508 deletion in the cystic fibrosis transmembrane conductance regulator gene was identified as previously described (10). Plasma 1-Antitrypsin Plasma 1-antitrypsin levels were measured in all participants with the EE (n = 46, ES [n = 40], EZ (n = 26), SS (n = 12), SZ (n = 10), or ZZ (n = 6) genotype and in randomly selected subgroups