Chronic Obstructive Pulmonary Disease•

Clinical features and 1-year outcomes of variable obstruction in participants with preserved spirometry: results from the ECOPD study in China

Abstract

Background There are limited data on the clinical features and longitudinal prognosis of variable obstruction, particularly among never smokers and different variable obstruction types. Therefore, we aimed to evaluate the clinical characteristics of the participants with variable obstruction and determine the relationship between variable obstruction and the development of chronic obstructive pulmonary disease (COPD) and the decline of lung function in a community-dwelling study of Chinese, especially among never smokers and different variable obstruction subtypes.

Methods Participants with preserved spirometry (postbronchodilator forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) ≥0.70) at baseline from the Early COPD cohort were included in our analysis. Participants with variable obstruction (prebronchodilator FEV1/FVC <0.70) were compared with those without variable obstruction (prebronchodilator FEV1/FVC ≥0.70). We performed subgroup analyses in never smokers, former and current smokers, and different variable obstruction types (postbronchodilator FVC <prebronchodilator FVC or postbronchodilator FVC ≥prebronchodilator FVC).

Results The final analysis included 1140 participants with preserved spirometry (169 in the variable obstruction group) at baseline. Participants with variable obstruction were older, had lower lung function and had greater severe emphysema and computed tomography-defined air trapping than participants without variable obstruction. Participants with variable obstruction had a significantly increased risk of incident spirometry-defined COPD (relative risk: 3.22, 95% confidence interval 2.23 to 4.64, p <0.001) than those without variable obstruction after adjustment for covariates. These findings remained consistent among both former and current smokers, never smokers, and different variable obstruction types. Additionally, participants with variable obstruction had a faster decline in postbronchodilator FEV1/FVC (2.3±0.5%/year vs 0.9±0.4%/year, mean difference: 1.4 (95% confidence interval 0.5 to 2.3), p=0.002) than participants without variable obstruction after adjustment for covariates.

Conclusions The results of our study revealed that variable obstruction can identify individuals who are at risk for the development of COPD and accelerated postbronchodilator FEV1/FVC decline in preserved spirometry.

What is already known on this topic

  • Recently, an analysis of the Subpopulations and Intermediate Outcome Measures in Chronic Obstructive Pulmonary Disease (COPD) Study has found that variable obstruction predicts future COPD development in smokers (≥20 pack years), suggesting that they may be a potential population for pre-COPD. However, there are limited data on the clinical features and longitudinal prognosis of variable obstruction, particularly among never smokers and different variable obstruction types.

What this study adds

  • Our results indicated that participants with variable obstruction had significant small airway dysfunction, worse lung function and more severe lung structural changes. Participants with variable obstruction also had a higher risk of COPD development and a faster decline rate in forced expiratory volume in 1 s/forced vital capacity. Similar results were also found in subgroups of smokers, never smokers and different variable obstruction types.

How this study might affect research, practice or policy

  • Individuals with variable obstruction are among those in the pre-COPD, and it is necessary to strengthen risk factor intervention, follow-up, management of these individuals and even explore necessary medicine treatment.

Introduction

Chronic obstructive pulmonary disease (COPD) is the third leading cause of mortality worldwide, and its prevalence is still rising.1 The complex natural history of COPD and the lack of identification of patients in the early process of the disease cause few treatment options beyond symptom control, which primarily drives the global economic burden of COPD continuously increase.2 Emerging studies are investigating early COPD to break this dilemma.3–7 However, there are currently no tools for accurately diagnosing early COPD,8 9 focusing on pre-COPD may be a more feasible approach.8–12

The fundamental abnormality in all patients with COPD is persistent airflow limitation, defined as postbronchodilator forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC)<0.70.9 Individuals with reversible airflow limitation (prebronchodilator FEV1/FVC <0.70 and postbronchodilator FEV1/FVC ≥0.70), also known as variable obstruction, do not meet the diagnosis of COPD.10 Individuals with variable obstruction have a similar presentation of clinical characteristics to patients with COPD. Recently, an analysis of the Subpopulations and Intermediate Outcome Measures in COPD Study (SPIROMICS) cohort has found that variable obstruction predicts future COPD development in smokers (≥20 pack years), suggesting that they may be a potential population for pre-COPD.10,1 The Global Burden of Disease Study 2017 shows that smoking (both active and passive) accounts for only about 50% of the total population attributable risk of COPD globally.1 The association between variable obstruction and COPD risk in the general population and never smokers remains unclear.

Physiologically, the contracted smooth muscle is relaxed after the use of bronchodilators, resulting in increased FEV1, leading to variable obstruction.13 However, some people’s FVC decreased after using bronchodilators,14 while in most people FVC is generally consistent before and after using bronchodilators, which can also lead to variable obstruction.15 Whether participants with variable obstruction from decreased postbronchodilator FVC are as susceptible to COPD as those from increased postbronchodilator FEV1 is unclear. To fill in the above knowledge gaps, we aim to explore the clinical features and respiratory outcomes of participant with variable obstruction in the general population. Meantime, we performed subgroup analyses according to smoking status and different variable obstruction types.

Methods

Study design and population

The Early Chronic Obstructive Pulmonary Disease (ECOPD) cohort (ChiCTR1900024643) is a multicentre, prospective, observational cohort study conducted in Guangdong Province, China. The study protocol, operational quality control criteria and participants recruitment for the ECOPD cohort study have been previously published.3 16 17 In short, participants first completed the COPD epidemiological questionnaire and spirometry in the community. All participants with spirometry results of FEV1/FVC <0.70 were recommended to further enter the ECOPD study to complete chest computed tomography (CT), impulse oscillometry (IOS), and other tests. For participants with spirometry results of FEV1/FVC ≥0.70, one out of every four spirometry will be selected according to the order of completion of spirometry and further recommended to enter the ECOPD study.

This study included all baseline data from the ECOPD cohort, completed between July 2019 and August 2021. This study also included 1-year follow-up data until August 2022 for prognostic analysis. The main inclusion criteria include (1) aged 40–80 years; (2) signed written informed consent; (3) completion of quality control questionnaires, prebronchodilator and postbronchodilator spirometry, and chest CT scans. The main exclusion criteria include acute respiratory infection or acute exacerbation of COPD within 4 weeks before enrollment.3 All participants signed written informed consent before enrollment.

Patient and public involvement

This was a prospective cohort study, and no patients were directly involved in our study design, setting the research questions or the outcome measures. No patients were asked to advise on interpretation or writing up of the results.

Questionnaires

The COPD epidemiological questionnaire was modified from the questionnaire used in the previous Burden of Obstructive Lung Disease study and the COPD epidemiological survey in China.18 19 Face-to-face interviews are conducted by well-trained and qualified staff. The COPD epidemiological questionnaire included demographic data, smoking status, smoking index, family history of respiratory diseases, biomass exposure history, and occupational dust exposure history. The acute respiratory event was defined as at least two symptoms of cough, sputum production, purulent sputum, wheezing, and dyspnoea lasting more than 48 hours, after ruling out the presence of cardiac insufficiency, pulmonary embolism, pneumothorax, pleural effusion, or cardiac arrhythmia.20 21 Mild, moderate and severe are further classified according to the treatment modality. The mild acute respiratory event was defined as requiring additional medical treatment and being able to be managed alone at home. The moderate acute respiratory event was defined as requiring outpatient or emergency care. The severe acute respiratory event was defined as requiring hospitalisation.

Lung function

The quality control standards and operating specifications of lung function were carried out in accordance with the recommendations of the American Thoracic Society and the European Respiratory Society in 2005.22 23 The predicted value of lung function was calculated using the 1993 predicted value formula of the European Community for Steel and Coal and multiplied by the Chinese conversion factor.24 25 Preserved spirometry was defined as postbronchodilator FEV1/FVC ≥0.70 at baseline.10 26–29 Variable obstruction was defined as postbronchodilator FEV1/FVC ≥0.70 but prebronchodilator FEV1/FVC <0.70.10 We further divided it into variable obstruction 1 and variable obstruction 2 according to the reasons for the formation of variable obstruction.29 Participants with postbronchodilator FVC <prebronchodilator FVC were defined as variable obstruction 1 and participants with postbronchodilator FVC ≥prebronchodilator FVC were defined as variable obstruction 2. We further used a lower limit of normal (LLN) to replace 0.70 to establish a new variable obstruction diagnostic method for sensitivity analysis. New definition of variable obstruction was defined as postbronchodilator FEV1/FVC ≥LLN and prebronchodilator FEV1/FVC <LLN.

Airflow reversibility was defined as an improvement in postbronchodilator FEV1 of more than 12% and 200 mL compared with prebronchodilator FEV1.30 Spirometry-defined small airway dysfunction was diagnosed by at least two of maximal mid-expiratory flow (MMEF), forced expiratory flow at the 50% FVC (FEF50), and forced expiratory flow at the 75% FVC (FEF75) being less than 65% of the predicted value.31 32 The development of airflow limitation was defined as participants with preserved spirometry at baseline and the postbronchodilator FEV1/FVC <0.70 at 1-year follow-up.9 The quality control standards and operating specifications of the IOS were carried out according to the recommendation of the European Respiratory Society in 2003.33 Small airway dysfunction (SAD) detected by IOS was defined as the difference between the resistance at 5 Hz versus the resistance at 20 Hz (R5–R20) >0.07 kPa/L/s at baseline.34 35

Imaging

High-resolution inspiratory and expiratory CT was performed with Siemens Definition AS Plus 128-slicers and United-imaging uCT 760 128-slicers.3 The inspiratory phase was measured close to total lung capacity (TLC), and the expiratory phase was measured close to residual volume (RV). To ensure the quality of CT scanning, we fully trained the subject to inhale and exhale before the operation, and also evaluated whether the scan results of the participant’s inspiratory and expiratory phases were satisfactory during the scan. The specific CT plan is described in the previous.3 Emphysema was quantified by the percentage of low-attenuation areas lower than −950 Hounsfield units in inspiratory CT. Air trapping was quantified by the percentage of low-attenuation areas lower than −856 Hounsfield units in expiratory CT.36

Statistical analysis

For baseline variables, the between-group difference was compared using a t-test, χ2 test, or Fisher’s exact test, as appropriate. Chronic respiratory symptom score, lung function, IOS, and chest CT continuous variable were compared between groups using multivariable linear regression analysis. The results of categorical variables of chronic respiratory symptom score, lung function, IOS, and chest CT were compared using multivariable logistic regression analysis. For respiratory health outcomes, we used log-binomial regression models to compare the differences in the risk of progression to airflow obstruction between the two groups after 1 year.37 38 We used analysis of covariance to compare the differences in the annual decline of lung function between the two groups. Poisson regression adjusted for overdispersion distributions was used to analyse the risk of acute respiratory events.39 40 Covariates were selected for items known to be associated with the outcomes of interest according to the prior literature. Adjusted confounding factors included age, sex, body mass index (BMI), smoking status, smoking index, occupational history of dusts/gases/fumes, biomass exposure history, and family history of respiratory diseases. When analysing the annual decline rate of lung function, the baseline lung function value was added on the basis of adjusting the above confounding factors. In addition, we conducted subgroup analysis according to different smoking status and VO types. All analyses were performed using IBM SPSS V.27.0 software. Two-sided p <0.05 was considered the difference to be statistically significant.

Results

The study flow chart is presented in figure 1. A total of 1140 participants with preserved spirometry from the ECOPD cohort were included in this study. Of 1140, 169 participants met the variable obstruction diagnostic criteria and were classified into the variable obstruction group, and 971 participants were classified into the never obstructed group.

Figure 1
Request permissionrequest permission icon
Figure 1

Study flow chart. ECOPD, Early Chronic Obstructive Pulmonary Disease; CT, computed tomography.

Clinical characteristics at baseline

Table 1 showed the baseline characteristics of the participants with preserved spirometry. Participants with variable obstruction were older (61.4±7.8 years old vs 58.0±7.7 years old, p <0.001) and had a higher proportion of men (75.1% vs 55.6 %, p <0.001), lower BMI, a higher proportion of current smokers (46.7% vs 30.7%, p <0.001), and higher smoking index (27.3±30.0 pack years vs 17.5±28.2 pack years, p<0.001) compared with those without variable obstruction. There were no significant differences between the two groups in biomass exposure, occupational exposure, family history of respiratory diseases, previous diagnosis of asthma, comorbidity, and acute respiratory events during the preceding year. Notably, the variable obstruction group had a higher proportion of previous diagnosis of COPD compared with the never obstructed group (4.1% vs 1.1%, p=0.004).

Table 1
|
Baseline characteristics of the participants with preserved spirometry in ECOPD study

Tables 2 and 3 show the chronic respiratory symptom, spirometry, IOS, and CT results of the participants with preserved spirometry at baseline. The mMRC dyspnoea score (0.26±0.49 vs 0.24±0.51, adjusted mean difference: 0.01 (95% CI −0.08 to 0.09)), the proportion of mMRC ≥2, the CAT score (3.2±4.2 vs 3.1±3.9, adjusted mean difference: 0.0 (95% CI −0.6 to 0.7)), and the proportion of CAT ≥10 were not significantly different between the variable obstruction and never obstructed groups. The variable obstruction group had worse prebronchodilator and postbronchodilator FEV1 and FEV1 % of predicted value, and FEV1/FVC compared with the never obstructed group. Participants with variable obstruction had larger prebronchodilator FVC and FVC % of predicted value than those without variable obstruction. However, no statistical significance in postbronchodilator FVC % of predicted value was found between those with and without variable obstruction.

Table 2
|
Baseline chronic respiratory symptom, spirometry, impulse oscillometry, and computed tomography results of the participants with preserved spirometry
Table 3
|
Baseline chronic respiratory symptom, spirometry, impulse oscillometry, and computed tomography results of the participants with preserved spirometry

Compared with participants without variable obstruction, the prebronchodilator and postbronchodilator spirometry parameters of SAD (MMEF, FEF50 and FEF75) in participants with variable obstruction were lower (p < 0.05 for all comparisons) after adjustment for covariates. The CT parameters of air trapping (RV/TLCCT and expiratory LAA-856) were significantly greater in variable obstruction than those without variable obstruction after adjustment for covariates. SAD-related IOS parameters (R5-R20, AX, and Fres) were significantly greater in the variable obstruction group (p <0.05 for all comparisons) but X5 had no difference compared with the never obstructed group after adjustment for covariates. Multivariable logistic regression showed that participants with variable obstruction possessed more postbronchodilator spirometry-defined SAD (adjusted odds ratio [OR]=10.18 (95% CI 5.84 to 17.74, p <0.001), CT-defined air trapping (adjusted OR=1.61 (95% CI 1.02 to 2.54), p=0.043), emphysema (adjusted OR=1.66 (95% CI 1.14 to 2.44), p=0.009) compared with those without variable obstruction after adjustment for covariates.

Variable obstruction and acute respiratory events

In the entire cohort, 88% (1008/1140) of the participants completed the 1-year follow-up for acute respiratory events. The risk of acute respiratory events in participants with preserved spirometry was shown in online supplemental table S1. The risk of total acute respiratory events (0.29±0.02 per patient year vs 0.31±0.17 per patient year, adjusted relative risk (RR)=1.01 (95% CI 0.73 to 1.39), p=0.975) and moderate-to-severe acute respiratory events (0.19±0.02 per patient year vs 0.19±0.05 per patient year, adjusted RR=1.02 (95% CI 0.68 to 1.53), p=0.942) in participants with variable obstruction did not increase significantly compared with participants without variable obstruction. Similar results were seen in never smokers, and former and current smokers.

Variable obstruction was associated with accelerated postbronchodilator FEV1/FVC decline

In the entire cohort, 80% (918/1140) of the participants completed the 1-year follow-up for spirometry. Tables 4 and 5 show the estimated annual declines in postbronchodilator spirometry. Participants with variable obstruction had a faster decline in postbronchodilator FEV1/FVC (2.3±0.5 %/year vs 0.9±0.4%/year, adjusted mean difference: 1.4 (95% CI 0.5 to 2.3), p=0.002) than participants without variable obstruction after adjustment for covariates. A similarly positive association was found in former and current smokers, participants with variable obstruction 2, and sensitivity analysis using LLN as the diagnostic cut-off value for FEV1/FVC (online supplemental table S2). Among never smokers, participants with variable obstruction had a higher annual decline rate in FEV1/FVC (1.8±0.8%/year vs 0.9±0.5%/year, adjusted mean difference: 0.9 (95% CI −0.6 to 2.3), p=0.224) but did not reach statistical significance. A similar situation was also found in participants with variable obstruction 1. There was no significant difference in the annual decline rates of FEV1 and FVC between the two groups.

Table 4
|
Estimated annual declines in postbronchodilator spirometry
Table 5
|
Estimated annual declines in postbronchodilator spirometry stratified by different variable obstruction types

Variable obstruction was associated with spirometry-defined COPD development

The transition of lung function categories from baseline to 1-year follow-up is shown in figures 2 and 3. Thirty percent (40/134) of participants with variable obstruction and 7% (54/784) of participants without variable obstruction had spirometry meeting the diagnostic criteria for COPD (postbronchodilator FEV1/FVC<0.70) at 1-year follow-up. The variable obstruction group had an increased risk for spirometry-defined COPD development compared with the never obstructed group (adjusted RR=3.22 (95% CI 2.23 to 4.64), p <0.001) after adjustment for covariates. The results remained robust in never smokers, former and current smokers, and sensitivity analysis using LLN as the diagnostic cut-off value for FEV1/FVC (figure 2 and online supplemental figure S1). When variable obstruction was further divided into variable obstruction 1 and variable obstruction 2, participants with variable obstruction 1 and variable obstruction 2 both had a higher risk of developing COPD than participants with never obstruction after adjustment (figure 3). These results were maintained among those who had never smoked or smoked.

Figure 2
Request permissionrequest permission icon
Figure 2

Transition of lung function categories from baseline to 1-year follow-up. COPD, chronic obstructive pulmonary disease; CI, confidence interval; RR, relative risk. Log-binomial regression models adjusted for age, sex, body mass index, smoking status, smoking index, biomass exposure, occupational history of dusts/gases/fumes, and family history of respiratory diseases.

Figure 3
Request permissionrequest permission icon
Figure 3

Transition of lung function categories from baseline to 1-year follow-up stratified by different variable obstruction types. COPD, chronic obstructive pulmonary disease; CI, confidence interval; FVC, forced vital capacity; RR, relative risk. Log-binomial regression models adjusted for age, sex, body mass index, smoking status, smoking index, biomass exposure, occupational history of dusts/gases/fumes, and family history of respiratory diseases.

Discussion

Our study demonstrated that variable obstruction was associated with a higher risk of progression to COPD and accelerated postbronchodilator FEV1/FVC decline compared with those without variable obstruction in preserved spirometry. These results were robust in never smokers, former and current smokers, and different types of variable obstruction.

This study is the first to explore the clinical characteristics and longitudinal prognosis of individuals with variable obstruction in a Chinese community population. Individuals with variable obstruction have a higher proportion of exposure to risk factors for COPD, more COPD-like lung structural changes, and small airway functional changes compared with those with never obstruction. Combined with the 1-year follow-up results, it is shown that variable obstruction is more likely to progress to COPD. Clinicians and lung function technicians need to pay more attention when encountering individuals with variable obstruction in clinical practice. Individuals with variable obstruction are among those in the pre-COPD,9 11 and it is necessary to strengthen risk factor intervention, follow-up and management of these individuals and even explore necessary medicine treatment. A recent result from the Tasmanian Longitudinal Health Study showed that prebronchodilator FEV1/FVC less than the 10th percentile can identify high-risk population for COPD. Postbronchodilator spirometry did not help improve the accuracy of identifying high-risk population for COPD.41 Combined with the findings of variable obstruction in this study, it may represent that prebronchodilator spirometry is adequate for COPD screening and identification of pre-COPD in clinical practice.

The results of this study are consistent with the SPIROMICS cohort study, where we found that participants with variable obstruction had worse lung function, more severe emphysema and air trapping.10 Our results for the whole population include never smokers and persons exposed to other risk factors for COPD such as biomass exposure, occupational dust exposure, and family history of respiratory diseases. Compared with the SPIROMICS cohort, which only includes heavy smokers, the results of our study have better extrapolation. The difference is that the proportion of airway reversibility (12% vs 31%) and previously diagnosed asthma (1% vs 11%) in participants with variable obstruction in our study were significantly lower than those in the SPIROMICS cohort. The reason for the difference may be the ECOPD cohort recruited subjects in the community, while the SPIROMICS cohort included subjects from the hospital.42 Results for other cohorts are currently unavailable. Further cohort studies with large sample sizes and long-term follow-up are still needed to further explore the prognosis of variable obstruction and the differences between different variable obstruction types.

Notably, we found that 91% of participants with variable obstruction had spirometry-defined SAD, which was significantly higher than that of participants without variable obstruction. Participants with variable obstruction also had lower MMEF, MMEF % of predicted value, FEF50, FFE50 % of predicted value, FEF75, FEF75 % of predicted value, and higher R5–R20 compared with those without variable obstruction. Previous studies have identified SAD as a pre-COPD.43 44 This study found that participants with variable obstruction had more severe SAD, which was assessed by IOS, CT or spirometry. Therefore, the results of this study further support that variable obstruction meets one of the definitions of pre-COPD.11

There are two ways to lead to postbronchodilator FEV1/FVC ≥0.70 for participants with prebronchodilator FEV1/FVC <0.70. One is an increase in FEV1 and the other is a decrease in FVC. Variable obstruction 1 (postbronchodilator FVC <prebronchodilator FVC) represents the FVC-decreasing type and variable obstruction 2 (postbronchodilator FVC ≥prebronchodilator FVC) represents the FEV1-increasing type in our study. Previous studies found that participants with paradoxical bronchodilator response (postbronchodilator FEV1<90% prebronchodilator FEV1) were associated with worse chronic respiratory symptoms and poor quality of life.27 45 It implies that the variable obstruction of the FVC-reduced type may be different from the variable obstruction of the FEV1-increased type. Our study found that variable obstruction 1 and variable obstruction 2 also had a higher risk of COPD and accelerated postbronchodilator FEV1/FVC decline. Collectively, although the clinical characteristics of different variable obstruction types may be different, they are all high-risk individuals for the development of COPD and should be paid attention.

The 1-year follow-up period in this study was too short for assessing the annual decline in lung function and the risk of COPD development.46 The development of airflow limitation may be due to its physiological variation.47 Therefore, the data on the annual decline in lung function and the development of COPD in this study need to be treated with caution. However, we found that participants with variable obstruction had a higher proportion of SAD and more severe emphysema at baseline. Previous studies have shown that SAD and emphysema in individuals with preserved spirometry were risk factors for accelerated lung function decline and the development of COPD.48–50 The baseline findings and 1-year longitudinal prognostic results of this study verified each other, so they still have certain reference value.

Our study had several limitations. First, the number of participants who met the diagnostic criteria for variable obstruction in the ECOPD cohort was limited, and more subgroup analyses could not be performed to further explore the clinical characteristics and longitudinal prognosis of different subgroups of variable obstruction. Second, the diagnosis of variable obstruction fluctuates greatly, and some participants with variable obstruction have normal spirometry after 1–3 years of follow-up.10 We only make a diagnosis based on whether the baseline spirometry of the participants meets the diagnostic criteria for variable obstruction, so the results of this study do not guide the follow-up and management of participants with fluctuating variable obstruction diagnoses. Further evaluation of the clinical characteristics and longitudinal outcomes of the participants with variable obstruction diagnosis fluctuation is needed in the future. Third, we included adjustments for some covariates that were found to potentially affect prognosis based on previous literature when we constructed the multivariable model. We did not use the directed acyclic graph to select covariates and there may be some confounding factors that potentially affect the results of this study that we did not measure and adjust. Fourth, about 15% of participants lost to follow-up during the 1-year follow-up, which may affect the analysis results of the prognosis. Finally, we did not exclude patients with previously diagnosed asthma by the physician and potentially patients with undiagnosed asthma. Asthma is a risk factor for the development and poor prognosis of COPD, which may affect the results of this study.51

Conclusion

Our results indicated that participants with variable obstruction had significant SAD, worse lung function, and more severe lung structural changes. Participants with variable obstruction also had a higher risk of COPD and a faster decline rate in FEV1/FVC. Similar results were also found in subgroups of former and current smokers, never smokers, and different variable obstruction types. Our results support the inclusion of variable obstruction as one of the definitions of pre-COPD and closer follow-up and management of variable obstruction. Further studies are needed to evaluate the effectiveness of risk factor intervention or pharmacal intervention on variable obstruction.

Overview

Article metrics
Altmetric data not available for this article.
Dimensionsopen-url