Rationale: A significant proportion of smokers have lung function impairment characterized by a reduced FEV1 with a preserved FEV1/FVC ratio. These smokers are a poorly characterized group due to their systematic exclusion from chronic obstructive pulmonary disease (COPD) studies.
Objectives: To characterize the clinical, functional, and radiographic features of Global Initiative for Chronic Obstructive Lung Disease (GOLD)-Unclassified (FEV1/FVC ≥ 0.7 and FEV1 < 80% predicted) and lower limits of normal (LLN)-unclassified (FEV1/FVC ≥ LLN and FEV1 < LLN) subjects compared to smokers with normal lung function and subjects with COPD.
Methods: Data from the first 2,500 subjects enrolled in the COPDGene study were analyzed. All subjects had 10 or more pack-years of smoking and were between the ages of 45 and 80 years. Multivariate regression models were constructed to determine the clinical and radiological variables associated with GOLD-Unclassified (GOLD-U) and LLN-Unclassified status. Separate multivariate regressions were performed in the subgroups of subjects with complete radiologic measurement variables available.
Measurements and Main Results: GOLD-U smokers account for 9% of smokers in COPDGene and have increased body mass index (BMI), a disproportionately reduced total lung capacity, and a higher proportion of nonwhite subjects and subjects with diabetes. GOLD-U subjects exhibit increased airway wall thickness compared to smoking control subjects and decreased gas trapping and bronchodilator responsiveness compared to subjects with COPD. When LLN criteria were used to define the “unclassified” group, African American subjects were no longer overrepresented. Both GOLD-U and LLN-Unclassified subjects demonstrated a wide range of lung function impairment, BMI, and percentage of total lung emphysema.
Conclusions: Subjects with reduced FEV1 and a preserved FEV1/FVC ratio are a heterogeneous group with significant symptoms and functional limitation who likely have a variety of underlying etiologies beyond increased BMI.
Clinical trial registered with www.clinicaltrials.gov (NCT000608764).
Smokers with reduced FEV1 or FVC in the setting of preserved FEV1/FVC ratio are a poorly characterized group due to their exclusion from most COPD studies.
Our study characterizes the clinical and radiographic features of these GOLD (Global Initiative for Obstructive Lung Disease)–unclassified smokers.
The heterogeneity of chronic obstructive lung disease (COPD) in clinical care has been compounded by, paradoxically, both underdiagnosis and over- or misdiagnosis. On the population level, massive underdiagnosis of obstructive lung disease exists and may contribute to undertreatment (1). However, among patients receiving medical care, the diagnosis of COPD is often made clinically, with only one-third of patients ever receiving confirmatory spirometry (2–5). Among patients who undergo spirometry, 8 to 14% (2, 4, 6) have evidence of impaired lung function that cannot be classified under current Global Initiative for Chronic Obstructive Lung Disease (GOLD) diagnostic and staging criteria (7). These GOLD-Unclassified (GOLD-U) subjects, who exhibit reduced FEV1 but maintain an FEV1/FVC ratio greater than or equal to 0.7, are a poorly characterized group due to their systematic exclusion from COPD studies.
Reduced FEV1 or FVC in the setting of a preserved FEV1/FVC ratio is often described as a restrictive pattern. However, substantial evidence has shown that this pattern on spirometry has poor predictive value for true restriction as assessed by reduced total lung capacity (TLC) (8–10). Despite lack of true restrictive disease, multiple studies have demonstrated that this pattern on spirometry is stable in the majority of subjects (11, 12) and is associated with increased mortality (13–15) and decreased functional status (16). Smokers with this spirometric pattern have been studied as subgroups in population-based studies, but extensive characterization of these subjects with regard to clinical and radiographic features has not been made previously. We hypothesized that subjects with the GOLD-U pattern of spirometry will have distinct clinical, functional, and radiographic characteristics when compared to smokers with normal lung function and subjects with COPD (GOLD stages 2–4).
Some of the results in the current manuscript have been previously reported in the form of an abstract (17).
COPDGene is an ongoing study that enrolled subjects from 21 clinical centers throughout the United States (clinicaltrials.gov identifier NCT000608764). Institutional review board approval was obtained at each participating clinical center, and all subjects provided informed consent. Details regarding the study design have been previously published (18). Subjects are self-identified non-Hispanic whites or African Americans with 10 or more pack-years of smoking, between the ages of 45 and 80 years. Subjects completed a modified American Thoracic Society Respiratory Epidemiology questionnaire, St. George's Respiratory Questionnaire, and 6-minute walk test. Pre- and post-bronchodilator spirometry was performed using an EasyOne spirometer (ndd, Zurich, Switzerland) according to the American Thoracic Society guidelines (19). Volumetric inspiratory and expiratory chest computed tomography (CT) scans using multidetector CT scanners were completed on all subjects per protocol (18).
FEV1 and FVC % predicted values and the lower limits of normal were calculated according to the prediction equations by Hankinson and colleagues (20). GOLD-U subjects have a post-bronchodilator FEV1 less than 80% predicted and FEV1/FVC greater than or equal to 0.7. Smoking control subjects have a post-bronchodilator FEV1/FVC greater than or equal to 0.7 and FEV1 greater than or equal to 80% predicted. Subjects with COPD have a post-bronchodilator FEV1/FVC less than 0.7 and FEV1 less than 80% predicted (GOLD stages 2–4). Bronchodilator response was considered positive if the subject had greater than or equal to 0.2-L increase in FEV1 or FVC and greater than or equal to 12% change in FEV1 or FVC above baseline (prebronchodilator) measurements. Additional variable definitions are provided in the online data supplement.
Quantitative CT measurements of total lung capacity (TLCCT), emphysema, and gas trapping percentages were performed using Slicer (Version 2, www.slicer.org). Percent emphysema was quantified as the percentage of lung volume on inspiratory CT with an attenuation less than −950 Hounsfield Units (HU). Gas trapping was quantified as the percentage of lung volume on expiratory CT (taken at functional residual capacity [FRC]) with an attenuation less than −856 HU. Airway analysis was performed using the Pulmonary Workstation Plus (VIDA Diagnostics, Inc., Coralville, IA) (21, 22). Details are provided in the online data supplement. The square root of wall area for a hypothetical airway with an internal perimeter of 10 mm (Pi10) was derived (23).
All analyses were performed using SAS (v9.1; Cary, NC). Comparisons between the GOLD-U group and smoking control subjects and between the GOLD-U group and subjects with COPD (GOLD stages 2–4) were made separately. Univariate comparisons were made using Fisher's exact test for dichotomous variables and a Student's t test or Wilcoxon rank sum test for normal and nonnormal continuous variables, respectively. Variables with a univariate P value less than 0.05 were considered as candidates in the multivariable regression model. Logistic regression with GOLD-U status as the dependent variable was performed using stepwise selection; all variables with a P value less than or equal to 0.05 were retained in the final model. Variables not retained as covariates in the final model were tested as confounders and retained in the model if greater than or equal to 10% change in the effect estimate was observed. Because radiographic variables were missing in some subjects due to technical limitations, a separate multivariable analysis was conducted in individuals with complete radiographic data. Selected analyses were repeated using lower limits of normal (LLN) (20) to define the smoking control subjects (LLN-control), COPD (LLN-COPD), and Unclassified (LLN-Unclassified) groups. Details are outlined in the online data supplement.
GOLD-U subjects represent 9.1% of the first 2,500 subjects recruited in COPDGene. Univariate comparisons between GOLD-U and smoking control subjects and GOLD-U and subjects with COPD are summarized in Table 1. Compared to smoking control subjects, GOLD-U subjects have greater pack-years of smoking, decreased 6-minute-walk distance (6MWD), lower resting oxygen saturation, greater subsegmental airway wall area, and increased rates of respiratory medication use and comorbid cardiovascular disease (such as congestive heart failure [CHF], hypertension, stroke, and transient ischemic attacks). Compared to subjects with COPD, GOLD-U subjects have fewer pack-years of smoking, higher rates of current smoking, greater 6MWD, reduced subsegmental airway wall area, and lower rates of bronchodilator responsiveness and respiratory medication use. Compared with both groups, GOLD-U subjects have higher body mass index (BMI), reduced TLC and emphysema, and a higher proportion of subjects with diabetes and who are of nonwhite race.
|GOLD-U Subjects||Smoking Control Subjects||Subjects with COPD|
|Age, yr||58.5 (8.7)||57.8 (8.9)||64.1 (8.4)*|
|Pack-years||43.7 (26.8)||37.5 (20.6)*||53.3 (26.7)*|
|Heart rate||76.8 (13.1)||74.3 (12.7)*||77.8 (13.1)|
|BMI||31.9 (7.2)||28.8 (5.9)*||28.0 (6.2)*|
|6-min walk distance, ft||1,251.7 (405.4)||1,491.8 (405.2)*||1,136.6 (429.1)*|
|Resting O2 saturation||96.3 (2.4)||97.0 (1.8)*||94.5 (3.6)*|
|FEV1, % predicted||70.4 (7.4)||97.9 (12)*||48.9 (18.1)*|
|FVC, % predicted||71.7 (8.3)||96.7 (12.1)*||76.3 (17.7)*|
|% Emphysema†||1.8 (2)||2.7 (3.0)*||15.5 (13.3)*|
|% Gas trapping‡||11.8 (8.5)||11.7 (9.5)||42.3 (20.3)*|
|Pi10§||3.8 (0.1)||3.7 (0.1)*||3.8 (0.1)|
|Subsegmental wall area %‖||65.0 (2.4)||63.0 (2.2)*||65.7 (2.4)*|
|TLC, % predicted||79.4 (13)||93.3 (15.1)*||102.6 (16.8)*|
|Functional residual capacity, % predicted||87.0 (18)||89.9 (20.1)||128.9 (33.4)*|
|Long-acting β-agonist, % of users||13.7||3.7*||51.7*|
|Long-acting muscarinic antagonist, % of users||8.4||2.3*||44.2*|
|Inhaled corticosteroids, % of users||15.9||4.7*||53.6*|
|Oral steroids, % of users||2.6||0.4*||6.2*|
|St. George's Respiratory Questionnaire||27.6 (23.0)||15.5 (17.4)*||40.3 (21.1)*|
|MMRC||1.4 (1.5)||0.7 (1.1)*||2.2 (1.4) *|
|Positive bronchodilator response||13.6||9.5||36.1*|
|History of CHF||5.3||0.8*||5.1|
|Obstructive sleep apnea||20.7||12*||16.6|
Because the spirometric pattern of reduced FEV1 in the setting of a preserved FEV1/FVC ratio is often described as being a restrictive pattern, we applied the prediction equations of Stocks and Quanjer (24) to the TLCCT (Table 2). Previous studies have demonstrated excellent correlation between CT-derived estimates of TLC and measurements obtained through helium dilution (25) and body plethysmography (26). Approximately one-half of the GOLD-U cohort was found to have restrictive lung disease as defined by a reduced TLCCT. Although the prevalence of restrictive lung disease is significantly higher in GOLD-U subjects, a substantial proportion of smoking control subjects and a small minority of subjects with COPD also have reduced TLCCT.
|GOLD-U Subjects||Smoking Control Subjects||Subjects with COPD|
|TLC,* L||4.5 (1.1)||5.3 (1.3)†||6.0 (1.4)†|
|TLC, % predicted||79.4 (13)||93.3 (15.1)†||102.6 (16.8)†|
|FVC < 80% predicted, % of subjects||88.6||6.0†||57.9†|
|TLC < 80% predicted, % of subjects||56.3||17.8†||9.0†|
|TLC < LLN, % of subjects||45.5||13.8†||7.2†|
The results of the multivariable analyses to determine the predictors of GOLD-U status are summarized in Table 3. In the models comparing GOLD-U to smoking control subjects, increased BMI, decreased 6MWD and resting oxygen saturation, higher subsegmental airway wall area, and history of CHF were significant and independent predictors of GOLD-U status. In the multivariate models comparing GOLD-U to patients with COPD, increased BMI and resting oxygen saturation, less emphysema, lower subsegmental airway wall area, and decreased rates of bronchodilator responsiveness were significant predictors. Decreased TLCCT was a significant predictor even after adjusting for BMI in analyses comparing GOLD-U subjects to smoking control subjects and subjects with COPD.
|GOLD-U versus Smoking Control Subjects|
|Resting O2 saturation||0.867||0.803–0.935|
|6MWD (per 100 ft)||0.890||0.857–0.924|
|Clinical + radiographic variables†|
|Resting O2 saturation||0.882||0.809–0.960|
|6MWD (per 100 ft)||0.948||0.905–0.993|
|History of CHF||3.460||1.107–10.817|
|Subsegmental wall area %||1.302||1.202–1.409|
|GOLD-U versus subjects with COPD (GOLD stages 2−4)|
|Age (per 10 yr)||0.630||0.490–0.810|
|Resting O2 saturation||1.244||1.149–1.347|
|6MWD (per 100 ft)||1.098||1.047–1.150|
|Clinical + radiographic variables¶|
|Resting O2 saturation||1.159||1.048–1.283|
|% Gas trapping||0.926||0.899–0.954|
|Subsegmental wall area %||0.774||0.693–0.865|
GOLD-U subjects are a heterogeneous group; the range of values of FEV1 % predicted, BMI, and percent emphysema is illustrated in Table 4. Subgroup analyses within the GOLD-U cohort by racial group and sex were conducted. There were no significant differences in age, current smoking status, or FEV1 % predicted by sex. Women with GOLD-U spirometry had reduced exercise capacity, significantly more dyspnea, and worse quality of life. Men with GOLD-U spirometry had significantly more pack-years of smoking and emphysema but reduced subsegmental airway wall area. Univariate comparisons by racial group are summarized in Table 5. African-American GOLD-U subjects were younger and had fewer pack-years of smoking, a higher proportion of current smokers, reduced exercise capacity, and a lower mean FEV1 % predicted than white subjects. African-American GOLD-U subjects also had less emphysema and thicker subsegmental airway wall measurements than white subjects. There were no differences in BMI by racial group or by sex.
|FEV1, % predicted||44–79|
|Age, yr||61.9 (8.8)||53.5 (5.6)*|
|Pack-years||48.4 (29.6)||36.9 (20.3)*|
|BMI||31.6 (6.9)||32.3 (7.8)|
|Current smoking, % of subjects||37.8||75.0*|
|FEV1 % predicted||71.3 (7.2)||69.2 (7.7)*|
|FEV1/FVC ratio||0.76 (0.05)||0.77 (0.05)|
|6MWD||1,333.1 (398.2)||1,132.1 (387.9)*|
|TLC, L (liters)||4.8 (1.1)||4.2 (1)*|
|% Emphysema||2.0 (2.2)||1.4 (1.5)*|
|% Gas trapping||11.5 (7.6)||12.1 (9.8)|
|Subsegmental wall area %||64.5 (2.1)||65.8 (2.4)*|
|Pi10||3.77 (0.10)||3.85 (0.14)*|
|Resting O2 saturation||95.9 (2.5)||96.9 (2.0)*|
|MMRC||1.2 (1.4)||1.6 (1.6)|
|SGRQ||26 (22.5)||29.9 (23.8)|
To further examine the role of BMI in GOLD-U subjects, univariate comparisons were made between obese (BMI ≥ 30) and nonobese (BMI < 30) subjects (Table 6). Obese GOLD-U subjects had significantly more dyspnea, reduced functional capacity as assessed by 6MWD, and lower resting oxygen saturation. BMI was significantly correlated with FEV1 % predicted (Rho = −0.21, P value = 0.0015) and subsegmental wall area measurements (Rho = 0.17, P value = 0.01) but was not significantly correlated with the percent of emphysema (P value = 0.13) or TLCCT (P value = 0.32).
|Age, yr||58.0 (8.8)||59.3 (8.6)|
|Sex, % male||46.6||50.0|
|Race, % African American||39.9||41.5|
|Pack-years||44.5 (27.7)||42.5 (25.7)|
|Current smoking (%)||49.6||57.5|
|FEV1 % predicted||70.0 (7.5)||71.0 (7.4)|
|FEV1/FVC ratio||0.77 (0.05)||0.76 (0.05)|
|6MWD||1,190.4 (412.1)||1,338.4 (381.3)*|
|% Emphysema||1.7 (1.7)||1.8 (2.3)|
|% Gas trapping||10.9 (6.2)||13.2 (11.0)|
|Subsegmental wall area %||65.1 (2.3)||64.9 (2.4)|
|Pi10||3.82 (0.13)||3.78 (0.11)*|
|Resting O2 saturation||95.8 (2.5)||96.9 (1.9)*|
|MMRC||1.7 (1.5)||1.0 (1.3)*|
|SGRQ||31.9 (24.2)||21.5 (19.9)*|
To examine the effects of multiple covariates on subsegmental airway wall area percentage in GOLD-U subjects, univariate analyses and multivariable regression were performed. Subsegmental airway wall area percentage demonstrated a significant negative correlation with FEV1 % predicted (Rho = −0.29, P value < 0.0001); the association remained significant after adjustment for age, sex, race, and pack-years smoked, as well as with and without adjustment for BMI. Subsegmental wall area percentage did not vary significantly by the presence or absence of bronchodilator response or chronic bronchitis.
Selected analyses were repeated using the LLN to define unclassified (LLN-Unclassified), smoking control (LLN-control), and COPD (LLN-COPD) groups. There were 13% fewer unclassified subjects, 16% more control subjects, and 10% fewer subjects with COPD in the LLN analysis. The concordance rate between GOLD-U and LLN-Unclassified groups was moderate (kappa = 0.72) (Table 7). LLN-Unclassified subjects are also a heterogeneous group (see Table E1 in the online supplement). Univariate comparisons between the LLN groups are summarized in Table E2. African American subjects account for a smaller proportion of LLN-Unclassified subjects compared to GOLD-U subjects. However, African American subjects in the LLN-Unclassified group continue to have lower FEV1 % predicted than white subjects despite being significantly younger and having fewer pack-years of smoking (Table E3). Differences by race in TLC, emphysema, and airway wall area percentage also persist. The prevalence of restrictive abnormalities in each of the LLN groups is similar to the prevalence when GOLD criteria are used (Table E4). In the multivariate models predicting LLN-Unclassified status, similar predictors were identified (Table E5). Increased BMI, resting oxygen saturation, and subsegmental airway wall area percentage are again identified as significant predictors. However, CHF and race are no longer significant predictors in any of the multivariate analyses.
Our study is the first to extensively characterize smokers with the GOLD-Unclassified pattern on spirometry with respect to clinical and radiographic variables. Although a considerable degree of heterogeneity within the GOLD-U group was noted, several distinguishing characteristics were identified.
The association of increased BMI in GOLD-U subjects relative to both smoking control subjects and subjects with COPD is consistent with previous reports in the literature (27). Increased body mass and adiposity are known to impact spirometry and lung volume measurements (28–30). Proportionately decreased FEV1 and FVC with resultant preservation of the FEV1/FVC ratio has been observed in overweight and obese subjects. However, the reductions in FEV1 and FVC, although statistically significant, are typically small, and FEV1 and FVC values usually remain well within the range of normal predicted values even in extreme obesity (29, 30). Thus, obesity alone is unlikely to account for the FEV1 impairment observed in GOLD-U subjects.
Moderate reductions in TLC have likewise been associated with increasing BMI, although once again, values typically remain within the range of normal (28). Among obese (BMI > 30) smoking control subjects in our cohort, the mean TLC % predicted was 91.3% (data not shown); thus we believe the degree to which TLC is reduced in the GOLD-U group is out of proportion to what would be expected from obesity alone. Interestingly, reduced FRC, which is often the most dramatic effect of obesity on lung function (28), is not observed in GOLD-U subjects relative to smoking control subjects. It should be noted that FRC measurements were acquired from CT data obtained in the supine position. FRC can be reduced significantly in normal subjects on assuming a recumbent position, whereas obese subjects, despite having a reduced FRC at baseline, typically experience no further reduction in FRC (31, 32). Despite these caveats, the FRC % predicted values for both groups remain within the normal range. Thus, although increased BMI clearly contributes to the respiratory impairment observed in some GOLD-U subjects, its role is neither comprehensive nor straightforward.
The association of low BMI with increased emphysema measured by quantitative CT has been well established in subjects with COPD (33, 34). The finding of reduced emphysema estimates in GOLD-U subjects who, on average, have higher BMI, would appear congruent with previous observations, although the mechanism behind this observation remains obscure. Reduced emphysema seems unlikely to be related to artifact from increased soft tissue mass, since this should cause increased image noise with resultant increase in the numbers of voxels below the −950 HU threshold. It is possible that this results from the relatively decreased TLCCT in these individuals, resulting in a relatively increased mean lung attenuation and a relative decrease in the percentage of voxels below the threshold value. In addition, the direction of causation between low BMI and increased emphysema has not been firmly established; severe emphysema can feasibly cause wasting, but it is less clear whether increased body mass would be protective against the development of emphysema.
Independent of BMI, an interesting feature of our GOLD-U cohort is the high proportion of African American subjects. The relative enrichment of African Americans in the GOLD-U cohort may represent an artifact caused by a combination of the spirometric prediction equations used for African American subjects and the fixed thresholds used to define GOLD-U subjects. The proportion of African American subjects relative to smoking control subjects is reduced when LLN criteria are applied. However, the consistently decreased proportion of African American subjects in the COPD cohort may reflect racial differences in the response to tobacco smoke. Analogous to studies in COPD populations (35), African Americans in the GOLD-U cohort were younger, had decreased cumulative exposure to tobacco smoke, and had lower FEV1 % predicted than white GOLD-U subjects.
Comorbid conditions identified as significant predictors of GOLD-U status include diabetes and CHF. Consistent with previous reports, diabetes appears to be a risk factor for impaired lung function independent of BMI (36, 37). Although the mechanism behind this association remains obscure, at least one study has demonstrated that pulmonary abnormalities may antedate the diagnosis of diabetes and suggests the two disorders may share a common origin (13, 36). Reduced FEV1 and FVC have also been associated with and may antedate the diagnosis of CHF (38–40). Interestingly, the self-reported rates of CHF were comparable in GOLD-U and subjects with COPD, even though obstruction is a relatively rare finding in CHF. This may reflect the diagnostic uncertainty frequently encountered in the clinical management of a dyspneic patient.
Increased subsegmental wall area percentage in GOLD-U subjects relative to smoking control subjects was an unexpected but consistent finding. In our analyses, subsegmental wall area percentage varied significantly by sex and race and was significantly associated with BMI. The literature regarding the association of wall area thickness and BMI has been inconsistent (33, 34); however, a positive correlation with segmental airway wall area percentage and BMI in subjects with COPD has been previously reported (33). Likewise, conflicting reports regarding differences in airway wall thickness by sex also exist. Several studies based on radiologic data have found increased wall area percentage in male subjects with COPD (41–43), whereas histological examination of lung tissue from subjects from the National Emphysema Treatment Trial found increased wall area percentage in female subjects (44). The increased subsegmental wall area percentage in female GOLD-U subjects may be due to processes distinct from those involved in obstructive lung disease. The lack of concurrent significant gas trapping and emphysema is consistent with this hypothesis. Whether increased subsegmental airway wall thickness represents a differential response to cigarette smoke in African Americans is unknown—to date, there have been no studies reporting airway wall thickness in African American subjects in either COPD or asthma.
Although aggregate measures are useful in characterizing GOLD-U subjects as a group, the heterogeneity of GOLD-U subjects becomes evident when the ranges of variables such as BMI, FEV1 % predicted, and % emphysema are examined. Multiple disease processes can be associated with the pulmonary function impairment observed in GOLD-U subjects and include interstitial lung disease (45), thoracic cage abnormalities, or functional impairments, such as diaphragmatic paralysis. It is also evident that a subset of GOLD-U subjects have processes typically associated with obstructive lung disease, such as emphysema. Longitudinal studies have demonstrated that approximately one-third of subjects with the GOLD-U pattern of respiratory impairment will develop obstructive lung disease (11). This group may benefit from early diagnosis and treatment. This is, however, contingent upon the ability to accurately diagnose these processes; although 21.6% of GOLD-U subjects and 28.3% of LLN-Unclassified subjects currently report a history of physician-diagnosed COPD (data not shown), the correlation with actual percent emphysema is poor and suggests considerable misclassification.
The current diagnostic and staging criteria systems for COPD, which are based solely on spirometry, do not address the overlap in underlying disease processes between GOLD-U, COPD, and smoking control subjects that likely exists. Many GOLD-U subjects have nontrivial amounts of emphysema, and a significant minority of smoking control subjects and subjects with COPD have evidence of reduced TLC. This diagnostic imprecision persists even when the groups are defined by LLN thresholds. Ideally, a diagnostic and staging system would allow for the identification of subjects with more than one concurrent disease process (46); such a system would necessarily be based on information from multiple arenas.
The strengths of the current study include the wealth of radiographic and epidemiological data on this previously poorly characterized group of smokers with GOLD-U spirometry. The limitations of our study include the possibility of spirometric artifacts, such as incomplete inspiration, which could theoretically account for the reduced FVC observed in GOLD-U subjects. And although COPDGene is more inclusive than most COPD studies, it is not a population-based study, and subjects are limited to former and current smokers; this may impact the generalizability of our findings. Although the sample size used in the current study has allowed for the identification of a considerable number of risk factors for the GOLD-U pattern of respiratory impairment, characterization of the subgroups within the GOLD-U population remains challenging. Subgroups of interest, which may represent differing pathophysiological processes, include subjects with emphysema versus subjects who are of African American race with increased airway wall thickness. Additional analyses on the remainder of the COPDGene cohort, as well as longitudinal and genetic analyses, will allow for more detailed studies in the future.
|1.||Mannino DM, Gagnon RC, Petty TL, Lydick E. Obstructive lung disease and low lung function in adults in the United States: data from the National Health and Nutrition Examination Survey, 1988–1994. Arch Intern Med 2000;160:1683–1689.|
|2.||Arne M, Lisspers K, Stallberg B, Boman G, Hedenstrom H, Janson C, Emtner M. How often is diagnosis of COPD confirmed with spirometry? Respir Med 2010;104:550–556.|
|3.||Han MK, Kim MG, Mardon R, Renner P, Sullivan S, Diette GB, Martinez FJ. Spirometry utilization for COPD: how do we measure up? Chest 2007;132:403–409.|
|4.||Kohler D, Fischer J, Raschke F, Schonhofer B. Usefulness of GOLD classification of COPD severity. Thorax 2003;58:825.|
|5.||Lee TA, Bartle B, Weiss KB. Spirometry use in clinical practice following diagnosis of COPD. Chest 2006;129:1509–1515.|
|6.||Zielinski J, Bednarek M, Gorecka D, Viegi G, Hurd SS, Fukuchi Y, Lai CK, Ran PX, Ko FW, Liu SM, et al.. Increasing COPD awareness. Eur Respir J 2006;27:833–852.|
|7.||Pauwels RA, Buist AS, Calverley PM, Jenkins CR, Hurd SS. 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;163:1256–1276.|
|8.||Aaron SD, Dales RE, Cardinal P. How accurate is spirometry at predicting restrictive pulmonary impairment? Chest 1999;115:869–873.|
|9.||Swanney MP, Beckert LE, Frampton CM, Wallace LA, Jensen RL, Crapo RO. Validity of the American Thoracic Society and other spirometric algorithms using FVC and forced expiratory volume at 6 s for predicting a reduced total lung capacity. Chest 2004;126:1861–1866.|
|10.||Vandevoorde J, Verbanck S, Schuermans D, Broekaert L, Devroey D, Kartounian J, Vincken W. Forced vital capacity and forced expiratory volume in six seconds as predictors of reduced total lung capacity. Eur Respir J 2008;31:391–395.|
|11.||Guerra S, Sherrill DL, Venker C, Ceccato CM, Halonen M, Martinez FD. Morbidity and mortality associated with the restrictive spirometric pattern: a longitudinal study. Thorax 2010;65:499–504.|
|12.||Iyer VN, Schroeder DR, Parker KO, Hyatt RE, Scanlon PD. The “non-specific” pulmonary function test: longitudinal follow up and outcomes. Chest 2010;139:878–886.|
|13.||Engstrom G, Hedblad B, Nilsson P, Wollmer P, Berglund G, Janzon L. Lung function, insulin resistance and incidence of cardiovascular disease: a longitudinal cohort study. J Intern Med 2003;253:574–581.|
|14.||Leone N, Courbon D, Thomas F, Bean K, Jego B, Leynaert B, Guize L, Zureik M. Lung function impairment and metabolic syndrome: the critical role of abdominal obesity. Am J Respir Crit Care Med 2009;179:509–516.|
|15.||Mannino DM, Buist AS, Petty TL, Enright PL, Redd SC. Lung function and mortality in the United States: data from the first National Health and Nutrition Examination Survey Follow-up Study. Thorax 2003;58:388–393.|
|16.||Mannino DM, Ford ES, Redd SC. Obstructive and restrictive lung disease and functional limitation: data from the third National Health and Nutrition Examination. J Intern Med 2003;254:540–547.|
|17.||Wan E, Hokanson J, Murphy J, Regan E, Make B, Lynch D, Crapo J, Silverman E. Clinical and radiographic predictors of GOLD-unclassified subjects in COPDgene. Am J Respir Crit Care Med 2011;183:A1010.|
|18.||Regan EA, Hokanson JE, Murphy JR, Make B, Lynch DA, Beaty TH, Curran-Everett D, Silverman EK, Crapo JD. Genetic epidemiology of COPD (COPDGene) study design. COPD 2010;7:32–43.|
|19.||Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, Crapo R, Enright P, van der Grinten CP, Gustafsson P, et al.. Standardisation of spirometry. Eur Respir J 2005;26:319–338.|
|20.||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;159:179–187.|
|21.||Aysola RS, Hoffman EA, Gierada D, Wenzel S, Cook-Granroth J, Tarsi J, Zheng J, Schechtman KB, Ramkumar TP, Cochran R, et al.. Airway remodeling measured by multidetector CT is increased in severe asthma and correlates with pathology. Chest 2008;134:1183–1191.|
|22.||Hoffman EA, Simon BA, McLennan G. State of the art. A structural and functional assessment of the lung via multidetector-row computed tomography: phenotyping chronic obstructive pulmonary disease. Proc Am Thorac Soc 2006;3:519–532.|
|23.||Nakano Y, Wong JC, de Jong PA, Buzatu L, Nagao T, Coxson HO, Elliott WM, Hogg JC, Pare PD. The prediction of small airway dimensions using computed tomography. Am J Respir Crit Care Med 2005;171:142–146.|
|24.||Stocks J, Quanjer PH. Reference values for residual volume, functional residual capacity and total lung capacity. ATS workshop on lung volume measurements. Official statement of the European Respiratory Society. Eur Respir J 1995;8:492–506.|
|25.||Iwano S, Okada T, Satake H, Naganawa S. 3D-CT volumetry of the lung using multidetector row CT: comparison with pulmonary function tests. Acad Radiol 2009;16:250–256.|
|26.||Brown MS, Kim HJ, Abtin F, Da Costa I, Pais R, Ahmad S, Angel E, Ni C, Kleerup EC, Gjertson DW, et al.. Reproducibility of lung and lobar volume measurements using computed tomography. Acad Radiol 2010;17:316–322.|
|27.||Mannino DM, Holguin F, Pavlin BI, Ferdinands JM. Risk factors for prevalence of and mortality related to restriction on spirometry: findings from the first National Health and Nutrition Examination Survey and Follow-up. Int J Tuberc Lung Dis 2005;9:613–621.|
|28.||Jones RL, Nzekwu MM. The effects of body mass index on lung volumes. Chest 2006;130:827–833.|
|29.||Schachter LM, Salome CM, Peat JK, Woolcock AJ. Obesity is a risk for asthma and wheeze but not airway hyperresponsiveness. Thorax 2001;56:4–8.|
|30.||Sin DD, Jones RL, Man SF. Obesity is a risk factor for dyspnea but not for airflow obstruction. Arch Intern Med 2002;162:1477–1481.|
|31.||Watson RA, Pride NB. Postural changes in lung volumes and respiratory resistance in subjects with obesity. J Appl Physiol 2005;98:512–517.|
|32.||Yap JC, Watson RA, Gilbey S, Pride NB. Effects of posture on respiratory mechanics in obesity. J Appl Physiol 1995;79:1199–1205.|
|33.||Mair G, Maclay J, Miller JJ, McAllister D, Connell M, Murchison JT, MacNee W. Airway dimensions in COPD: relationships with clinical variables. Respir Med 2010;104:1683–1690.|
|34.||Ogawa E, Nakano Y, Ohara T, Muro S, Hirai T, Sato S, Sakai H, Tsukino M, Kinose D, Nishioka M, et al.. Body mass index in male patients with COPD: correlation with low attenuation areas on CT. Thorax 2009;64:20–25.|
|35.||Dransfield MT, Davis JJ, Gerald LB, Bailey WC. Racial and gender differences in susceptibility to tobacco smoke among patients with chronic obstructive pulmonary disease. Respir Med 2006;100:1110–1116.|
|36.||Litonjua AA, Lazarus R, Sparrow D, Demolles D, Weiss ST. Lung function in type 2 diabetes: the Normative Aging Study. Respir Med 2005;99:1583–1590.|
|37.||van den Borst B, Gosker HR, Zeegers MP, Schols AM. Pulmonary function in diabetes: a metaanalysis. Chest 2010;138:393–406.|
|38.||Engstrom G, Melander O, Hedblad B. Population-based study of lung function and incidence of heart failure hospitalisations. Thorax 2010;65:633–638.|
|39.||Naum CC, Sciurba FC, Rogers RM. Pulmonary function abnormalities in chronic severe cardiomyopathy preceding cardiac transplantation. Am Rev Respir Dis 1992;145:1334–1338.|
|40.||Wright RS, Levine MS, Bellamy PE, Simmons MS, Batra P, Stevenson LW, Walden JA, Laks H, Tashkin DP. Ventilatory and diffusion abnormalities in potential heart transplant recipients. Chest 1990;98:816–820.|
|41.||Camp PG, Coxson HO, Levy RD, Pillai SG, Anderson W, Vestbo J, Kennedy SM, Silverman EK, Lomas DA, Pare PD. Sex differences in emphysema and airway disease in smokers. Chest 2009;136:1480–1488.|
|42.||Grydeland TB, Dirksen A, Coxson HO, Pillai SG, Sharma S, Eide GE, Gulsvik A, Bakke PS. Quantitative computed tomography: emphysema and airway wall thickness by sex, age and smoking. Eur Respir J 2009;34:858–865.|
|43.||Kim WJ, Silverman EK, Hoffman E, Criner GJ, Mosenifar Z, Sciurba FC, Make BJ, Carey V, Estepar RS, Diaz A, et al.. CT metrics of airway disease and emphysema in severe COPD. Chest 2009;136:396–404.|
|44.||Martinez FJ, Curtis JL, Sciurba F, Mumford J, Giardino ND, Weinmann G, Kazerooni E, Murray S, Criner GJ, Sin DD, et al.. Sex differences in severe pulmonary emphysema. Am J Respir Crit Care Med 2007;176:243–252.|
|45.||Washko GR, Hunninghake GM, Fernandez IE, Nishino M, Okajima Y, Yamashiro T, Ross JC, Estepar RS, Lynch DA, Brehm JM, et al.. Lung volumes and emphysema in smokers with interstitial lung abnormalities. N Engl J Med 2011;364:897–906.|
|46.||Bon JM, Weissfeld JL, Sciurba FC. A unique spirometric phenotype in COPD. Am J Respir Crit Care Med 2008;178:431–432.|
Supported by National Institutes of Health grants U01 HL089856 (to E.K.S.), U01 HL089897 (to J.D.C.), and T32HL007427 (to S.T.W.).
Author contributions: E.S.W. participated in data analysis and manuscript writing. J.E.H. participated in data collection, statistical support, and manuscript review. E.A.R. participated in data collection and manuscript editing. J.R.M. and B.J.M. participated in concept and design, data collection, and manuscript review. D.A.L. participated in data analysis and manuscript editing. J.D.C. and E.K.S. participated in funding support, concept and design, data collection, and manuscript review.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1164/rccm.201101-0021OC on April 14, 2011