Association between smoking cessation and obstructive spirometry pattern among Korean adults aged 40–79 years

Smoking cessation aids in restoring lung function. However, whether long-term cessation can fully restore lung function has not been studied thoroughly, especially in Asian countries. This study aimed to evaluate the association between smoking cessation status and obstructive spirometry pattern among Koreans aged 40–79 years. In total, 6298 men and 8088 women aged 40–79 years from the Korea National Health and Nutrition Examination Survey (2015–2019) were analyzed for smoking cessation status, including the duration after quitting. Current-smokers showed a higher likelihood of having an obstructive spirometry pattern than never-smokers among both men (odds ratio [OR]: 3.15, 95% confidence interval [CI]: 2.32–4.29) and women (OR: 2.60, 95% CI: 1.59–4.23). In men, the effect tended to decrease with longer duration after cessation, but male ex-smokers who had quit smoking ≥ 20 years ago still showed a higher likelihood of having an obstructive spirometry pattern than male never-smokers (OR: 1.40, 95% CI: 1.05–1.89). In female ex-smokers, there was no significant association with the obstructive spirometry pattern, compared to that in female never-smokers. This study emphasizes the benefits of smoking cessation, possibility of long-lasting harm to lung function due to tobacco smoking, and importance of smoking prevention.

www.nature.com/scientificreports/ Therefore, this study aimed to investigate the relationship between smoking cessation status, including the duration after cessation, and obstructive spirometry pattern among Korean adults aged 40-79 years, using a nationally representative survey. We also performed an additional analysis to investigate whether the cumulative smoking exposure or adolescent-onset smoking affected the association between lung function and smoking cessation status among male participants. Lastly, to minimize age differences across categories of smoking cessation status, separate analyses were performed after dividing the study participants into 10-year age groups.
Current-smokers showed a higher likelihood of having an obstructive spirometry pattern than never-smokers among both men (odds ratio [ www.nature.com/scientificreports/ ex-smokers showed no significant association with the obstructive spirometry pattern, when compared to female never-smokers. Other factors associated with obstructive spirometry pattern were greater pack-years of smoking (only in men), older age, residing metropolitan or rural area, physical inactivity (only in men), and higher height (Table 2). Similarly, FEV 1 and FEV 1 /FVC values showed an increasing tendency with longer duration of smoking cessation (Supplementary Table S1).
Compared to male never-smokers, male current-smokers who had started smoking before and after the age of 15 years showed ORs of 4.55 (95% CI: 2.08-9.94) and 3.16 (95% CI: 2.32-4.30), respectively, for the obstructive spirometry pattern. Male ex-smokers who had started smoking before and after the age of 15 years and had quit  (Fig. 2). Among men, the relationship between smoking cessation status and obstructive spirometry pattern was generally similar across age groups, except in the 40-49 years age group. Men aged 40-49 years did not show any significant association between smoking cessation status and obstructive spirometry pattern (Table 3).

Pack-years (< 20)
Pack-years (20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) Pack-years (≥ 30) Figure 1. Association between smoking cessation status and obstructive spirometry pattern according to the cumulative smoking exposure in men. Adjusted for age, educational level, household income, occupation, residential area, physical activity, high-risk drinking, and height. www.nature.com/scientificreports/ Since our study was cross-sectional design, the partial restoration of lung function observed can be explained in two ways. First, although the rate of lung function decline normalized, the decrease in maximal lung function due to smoking was not fully restored. Second, acceleration of the age-related decline in lung function due to smoking could not be fully normalized even after long-term cessation. Our results are consistent with recent cohort study that although smoking cessation has a benefit in terms of lung function, lung injury from smoking could persist for decades after smoking cessation 18 .
Tobacco smoking is related to pathophysiologic abnormalities of the lung, including inflammation 19-21 , immune dysfunction and increased susceptibility to infection [22][23][24] , mucus hypersecretion 25,26 , genetic abnormalities 27,28 , and airway remodeling [29][30][31] . Although such lung abnormalities can improve with smoking cessation, there is evidence for sustained pathophysiological abnormalities in ex-smokers 19,21 . Lung injuries during the active smoking period, such as irreversible emphysematous change, may also contribute to persistent lung function impairment.
Findings regarding heavy smokers or adolescent-onset smokers are also notable. In male current-smokers, greater cumulative smoking exposure and adolescent-onset smoking showed relatively higher effect sizes (represented as ORs) for the obstructive spirometry pattern than their counterparts. However, among male ex-smokers with a smoking cessation duration of ≥ 20 years, participants showed similar effect sizes for the obstructive spirometry pattern across categories of cumulative smoking exposure and age at smoking onset. Therefore, long-term smoking cessation may be more beneficial for high-risk smokers than for low-risk smokers, in terms of recovering lung function. These findings reinforce prior research that heavy smokers benefited from smoking cessation more than did light smokers, in the first year after cessation 13 .
This study has several limitations. First, the number of years of smoking cessation and cumulative smoking exposure were indirectly calculated without considering intermittent smoking history. Although self-reported smoking history is known to be highly reliable, the gap between the actual and estimated smoking history may have affected our results. Second, Korean women are likely to underreport their history of smoking owing to social unacceptance. Therefore, the number of female smokers was lower than expected in this study, and the results may have been distorted. Third, with the cross-sectional design, we could not estimate individuals' agerelated decline in lung function. Instead, the likelihood of having an obstructive spirometry pattern was used to evaluate the lung function of participants. The possibility of unmeasured confounding due to factors such as passive smoking, occupational exposure to harmful particles, or individual variation in metabolic enzyme activity should also be considered 32,33 . Thus, our results should be cautiously interpreted, when compared to the findings of similar studies. Further long-term prospective study, including accurate smoking history of women participants, is needed to evaluate impacts of smoking cessation on age-related decline in lung functions in Koreans.
Despite these limitations, this study has a major strength: this study was based on one of the most representative health statistics of the Korean population and a reliable measurement of lung function.
In conclusion, our study showed that a longer duration of smoking cessation was linked to a decreasing tendency in the likelihood of having an obstructive spirometry pattern among men. Our findings suggest that tobacco smoking causes long-lasting harm to lung function and indicates the importance of the prevention and cessation of smoking, particularly in high-risk male smokers. Table 3. Association between smoking cessation status and obstructive spirometry pattern according to age group a . OR odds ratio, CI confidence interval. a Adjusted for all covariates (cumulative smoking exposure, age, educational level, household income, occupation, residential area, physical activity, high-risk drinking, and height). b Obstructive spirometry pattern was defined as an FEV 1 /FVC < 0.7. The study included participants aged 40-79 years, which is the target population of the spirometry test in the KNHANES. Participants with asthma, with unreadable spirometry results, or who were unable to undergo the test were excluded from the study. Those who had incomplete or missing data were also excluded. The final study population included 6298 men and 8088 women.
Using pre-bronchodilator spirometry, lung function was measured at least two times for each participant and the largest value was reported.
Spirometry was performed by trained technicians, using the American Thoracic Society (ATS) /European Respiratory Society (ERS) 2005 standards 36,37 . The study included only valid spirometry results that met ATS/ ERS acceptability and repeatability criteria: (1) two or more spirometry curves should be free from artefacts, have good starts and show at least 6 s of exhalation; (2) two largest values of FEV 1 or FVC should be within 150 mL of each other 36 . The quality of the spirometry test was managed by the Korean Academy of Tuberculosis and Respiratory Diseases 38 .
Participants were classified as never-smokers, ex-smokers, or current-smokers according to their self-reported smoking status. Ex-smokers were subdivided into five categories according to the duration after smoking cessation: < 5 years, 5-10 years, 10-15 years, 15-20 years, and ≥ 20 years.
The covariates were selected based on the previous literature. Pack-years of smoking (< 20 years, 20-30 years, > 30 years), age (continuous variable), height (continuous variable) 39 , high-risk drinking (yes, no) 40,41 , physical activity (active, inactive) 42 , and socioeconomic factors such as education level (middle school or below, high school, college or above), household income (quartile of household income according to the 2015-2019 KNHANES survey), occupation (white collar, pink collar, blue collar, none or homemaker), area of residence (capital, metropolitan, rural) were included as covariates 43,44 . High-risk drinking was defined as having more than seven (men) or five (women) drinks at one time, at least twice per week 45,46 . Physical activity was evaluated using Global Physical Activity Questionnaire (GPAQ), developed by World Health Organization 47,48 . Statistical analyses. All analyses were performed separately by gender, considering gender-specific differences in smoking rate and smoking effect on lung function 49,50 . We compared the prevalence of obstructive spirometry pattern in current-smokers, in ex-smokers with various duration of cessation to that in never-smokers.
Chi-square tests (for categorical variables) and one-way analysis of variance (for continuous variables) were performed to determine differences in general characteristics between participants according to their smoking cessation status. Multiple logistic regression analysis was performed to calculate ORs with 95% CIs for evaluating the relationship between smoking cessation status and obstructive spirometry pattern. Additionally, multiple linear regression analyses were conducted to identify whether smoking cessation status was related to FEV 1 , FVC, or FEV 1 /FVC.
The data analysis for this paper was generated using SAS software, Version 9.4 of the SAS System for Unix. Copyright © 2016 SAS Institute Inc. SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc., Cary, NC, USA.