Chronic obstructive pulmonary disease (COPD) is a respiratory disease characterized by an irreversible limitation in airflow.¹⁾ The high prevalence and mortality rate of COPD has imposed a significant economic burden on both patients and health care delivery systems.²⁾ According to the World Health Organization statistics, there were 210 million people with COPD worldwide in 2000, and it was the fifth leading cause of death in 2002.³⁾ The fifth Korea National Health and Nutrition Examination Survey, conducted between 2010 and 2012 in South Korea, showed that the prevalence of COPD increased during each of the 3 years of the study period, with a mean prevalence of 13.7% among those surveyed.^4,5,6) Furthermore, lung diseases, including COPD, were the seventh leading cause of death in Korea.^4,5,6)

The risk factors of COPD include smoking, airway hyperresponsiveness, occupational exposures, and genetic diseases such as α₁ antitrypsin deficiency.¹⁾ Among these factors, smoking is estimated to be the primary causal factor for 80% of COPD-related deaths.⁷⁾ Smoking behavior and addiction can be analyzed using the Fagerstrom Test for Nicotine Dependence (FTND) that takes into account the time to first cigarette of the day (TTFC). The TTFC is the first item in the FTND and is known to reflect nicotine addiction and the ability to quit smoking.⁸⁾ A recent study on the relationship between COPD and TTFC concluded that a shorter TTFC correlates with an increased risk of COPD.⁹⁾ However, COPD was self-reported by the subjects in that study, and neither that nor other studies have used pulmonary function testing (PFT). PFT is an important measure for diagnosing obstructive lung disease, but clinical guidelines advise against screening asymptomatic patients for COPD owing to the associated costs.^1,10) Consequently, asymptomatic smokers can be easily overlooked. However, a single question regarding a smoker's TTFC would be cost-effective if shown to be a good screening method for smokers at risk of COPD.⁹⁾

Therefore, this study evaluated the relationship between TTFC and obstructive pulmonary disease, especially COPD, as measured by using spirometry in current smokers who visited Hallym University Sacred Heart Hospital.

The baseline characteristics of the participants are shown in detail in Table 3. In this study, we divided the subjects into two groups according to TTFC (early TTFC: ≤30 minutes after waking; late TTFC: >30 minutes after waking). About 25% of the current smokers were classified into the early TTFC group. Their mean age was 45.3 years. Significant differences were found in obstructive pulmonary impairment, FEV1/FVC, age, height, smoking behaviors (CPD and smoking duration), and education between the early TTFC and late TTFC groups (P<0.05). Compared with the late TTFC group, the early TTFC group had a higher age and lower education level.

Table 1 shows the results of the bivariate and multivariate logistic regression analyses adjusted for smoking behaviors (CPD and smoking duration), SES (education and income), age, and physical activity. At the bivariate (unadjusted; model 1) level, early TTFC was significantly related to obstructive pulmonary impairment (odds ratio [OR], 2.72; 95% confidence interval [CI], 1.29–5.74). The association between early TTFC and obstructive pulmonary impairment was similar in the model adjusted for smoking behaviors (CPD and smoking duration; model 2) and in the fully adjusted model (smoking behaviors [CPD and smoking duration], SES [education and income], age, and physical activity; model 3) at a statistically significant level (adjusted OR, 2.90; 95% CI, 1.37–6.17 for model 2; adjusted OR, 2.84; 95% CI, 1.22–6.61 for model 3 versus late TTFC). Table 2 shows the results of the bivariate and multivariate linear regression analysis of FVC%, FEV1%, and FEV1/FVC according to TTFC. TTFC was not significantly related to FVC%, FEV1%, and FEV1/FVC except at the bivariate level for FEV1/FVC (unadjusted coefficient estimate, −1.907; standard error, 1.790). However, the frequent negative values for the unadjusted coefficient estimate indicated that early TTFC could be associated with decreased pulmonary function overall.

Among current smokers in the present study, early TTFC was significantly associated with a higher risk of obstructive pulmonary impairment. In addition, considering several factors such as disease prevalence, age, and smoking behaviors, obstructive pulmonary impairment can be regarded as COPD in the present study.^1,4,5,6) The relationship between TTFC and COPD was statistically significant after adjusting for several confounders. Several reasonable mechanisms might explain the relationship between early TTFC and a higher risk of COPD. For instance, early TTFC smokers are prone to take more and larger puffs; therefore, more nicotine may accumulate in early TTFC smokers than in late TTFC smokers, concomitantly leading to a higher dose of carcinogens present in tobacco.¹⁶⁾ In addition, the presence of some genes increases the risk for COPD and nicotine dependence. For example, single nucleotide polymorphisms of the 15q25 locus of the a-nicotinic acetylcholine receptor (CHRNA 3/5) have been shown to simultaneously increase the risk of nicotine dependence and COPD, independent of smoking behaviors.^17,18) However, other genetic variations have been shown to be effected by smoking behaviors. For instance, CYP2A6, a nicotine-metabolism enzyme, is associated with both nicotine dependence¹⁹⁾ and COPD,²⁰⁾ and its activity is mediated by smoking behaviors.²¹⁾ Additional studies on the influence of genes and smoking behaviors on nicotine dependence and the development of COPD are necessary.

In the present study, some variables did not have a statistically significant association with TTFC. However, the CIs show the presence of similar trends between some variables and TTFC in the bivariate model. For example, just as nicotine dependence is associated with low SES,²²⁾ in the present analysis, the early TTFC group had similar CI trends in association with SES parameters such as lower income and education levels, even though the association between income and early TTFC was not statistically significant (P=0.272). This may have occurred because our defined cutoff income values did not accurately reflect the different levels of SES.¹³⁾

Meanwhile, smoking and obesity are significant causes of mortality and morbidity worldwide.²³⁾ However, the association between nicotine dependence and obesity has been controversial. Weight loss could be associated with nicotine dependence because of an increased metabolic rate and decreased caloric absorption resulting from a reduction in appetite.²⁴⁾ However, heavy smokers may gain weight because they are more prone to adopt behaviors leading to weight gain than are nonsmokers or light smokers. Additionally, they consume fewer vegetables, adopt non-nutritious food intake, drink more alcohol, and have a tendency for less physical activity.^25,26) Our study also showed no significant association between TTFC and obesity (P-values: weight, 0.062; BMI, 0.875; and body fat percentage, 0.556). Therefore, further studies are needed to more precisely define the relationship between obesity and TTFC.

PFT is an important method for the diagnosis of COPD, but it has largely been excluded from previous cohort studies on the association between TTFC and COPD.⁹⁾ To our knowledge, the present study is the first to investigate the association between TTFC and PFT by using spirometry. Several reports have suggested an association between smoking behaviors and PFT. Greater addiction to nicotine in smokers is associated with a higher risk of obstructive lung disease.²⁷⁾ Higher FTND scores reflect greater nicotine dependence among smokers with COPD than among smokers without COPD²⁸⁾ and ex-smokers who quit smoking after a COPD diagnosis.²⁷⁾ The relationship between smoking and poor PFT values was highly significant. Smokers had a 17.3-times greater risk of having impaired pulmonary functions than did nonsmokers.²⁹⁾ The association between respiratory symptoms and smoking duration is statistically significant.³⁰⁾ Significant negative relationships were observed between the duration and amount of smoking and PFT values.³¹⁾ In a similar vein, this study also showed that early TTFC was associated with obstructive pulmonary impairment by itself after adjusting for CPD and smoking duration (Table 1). Moreover, our model 3 (adjusted for smoking behaviors, age, education, income, and physical activity) revealed an association between early TTFC and obstructive pulmonary impairment. These data, including our findings, suggest that TTFC is a significant predictor of lung disease, especially COPD. Therefore, a single question regarding a smoker's TTFC may be a useful and cost-effective tool if it is verified to be an accurate screening method for smokers at risk of COPD.⁹⁾ Further study of smoking-related behaviors, especially TTFC, may help direct clinical screening for COPD, a serious disease and public health burden.

This study has several limitations. COPD should be diagnosed clinically in any patient who has a postbronchodilator FEV1/FVC <0.7 by using spirometry.¹⁾ Although the PFT data indicated obstructive respiratory impairment in the present study, it did not establish a diagnosis of COPD because we did not assess the response to bronchodilators in subjects with obstructive respiratory impairment. Therefore we could not exclude asthma, asthma–COPD overlap syndrome, bronchiectasis, and tuberculosis.¹⁾ However, given the disease prevalence, age, and smoking behaviors, COPD was the most likely cause of obstructive pulmonary impairment in the present study.^1,4,5,6) The prevalence of obstructive pulmonary impairment in the present study was lower than the prevalence of COPD in Korea. This may be because the mean age of our subjects was about 45 years, and this relatively young age may have limited the degree of respiratory impairment caused by smoking. ¹⁾ In addition, subjects who came to the medical center for a check-up may have had a healthier lifestyle. We also excluded subjects with restrictive respiratory impairment on PFT because of their small number. Moreover, the association between TTFC, FEV1%, and FVC% did not show statistical significance in the univariate and bivariate analyses. In the present study, only nine subjects had an FEV1 below 80% (COPD stage ≥2) (data not shown), and this may have affected the analysis of any association of TTFC with PFT values (FVC%, FEV1%, and FEV1/FVC).

Because of these reasons, additional larger studies, including PFTs with bronchodilators and TTFC, are needed to evaluate the relationship between lung function and nicotine dependence. Furthermore, large studies and cohort studies on patients of both sexes from a broader age range and varied ethnicities are needed because a greater understanding of smoking behaviors and COPD may allow earlier diagnosis and help develop improved strategies to limit disease progression.