Habitual Tea Drinking is Associated with Reduced Odds of Restrictive Spirometry Pattern in Non- smokers

Cheng-Yuan Hsu National Cheng Kung University Hospital I-Hsuan Wu National Cheng Kung University Medical Center: National Cheng Kung University Hospital Chieh-Ying Chou National Cheng Kung University Medical Center: National Cheng Kung University Hospital Yin-Fan Chang National Cheng Kung University Medical Center: National Cheng Kung University Hospital Zih-Jie Sun National Cheng Kung University Medical Center: National Cheng Kung University Hospital Yi-Ching Yang National Cheng Kung University Medical Center: National Cheng Kung University Hospital Feng-Hwa Lu National Cheng Kung University Medical Center: National Cheng Kung University Hospital Chih-Jen Chang National Cheng Kung University Medical Center: National Cheng Kung University Hospital Jin-Shang Wu (  jins@mail.ncku.edu.tw ) Department of Family Medicine, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan


Conclusion
Habitual tea drinking and tea consumption of at least three cup-years were associated with a reduced risk of RSP in the non-smokers, but not in the smokers.

Background
Pulmonary disorders, such as low forced vital capacity (FVC), forced expiratory volume in 1 second (FEV1), and FEV1/FVC are signi cant predictors of all-cause mortality [1]. Among these pulmonary impairments, obstructive lung diseases are the most well-recognized and contribute to most of the morbidity and mortality related to pulmonary impairments [2]. Restrictive spirometry pattern (RSP), which features with reduced FVC in the absence of air ow obstruction, has been shown to be as prevalent as chronic obstructive pulmonary disease (COPD) [3] as well as to be related to higher risks of all-cause and cardiovascular mortality [4,5]. Multiple clinical factors have been associated with RSP, including age, female sex, white race, lower education, abdominal obesity [6], ex-and current smoking [7], low physical activity level [8] and comorbidities including diabetes [9,10], metabolic syndrome [11,12], hypertension [10], stroke [13] and cardiovascular disease [14]. Careful management of these clinical conditions may be one way to prevent RSP.
Tea is a worldwide popular beverage, and tea consumption has been shown to lead to a reduced risk of cardiovascular disease, hypertension, diabetes, and lung cancer [15][16][17][18]. Twice daily tea consumption at a minimum has been found to be related to a lower risk of COPD [19]. As for subjects with COPD, smokers, especially those smoking over 20 pack-years, have been shown to have a higher prevalence of RSP compared with non-smokers [7]. However, little is known about the in uence of tea consumption on RSP.
Smoking has an impact on COPD and RSP, but the study of tea on lung function has been limited in nonsmokers. This study was aimed toward an investigation of whether tea consumption is associated with RSP in adults based on smoking status.

Subjects
This study recruited 19,637 participants, aged 20 years or more, who underwent a health examination at National Cheng Kung University Hospital from October 2001 to August 2009. Subjects with 1) a history of asthma, lung cancer, tuberculosis, or any pulmonary structural deformities; 2) connective tissue diseases such as rheumatoid arthritis, systemic lupus erythematosus, ankylosing spondylitis; 3) associated clinical missing data were excluded. Finally, a total of 14,997 subjects (8,805 men and 6,192 women) with ages ranging from 20 to 97 years were included in the analysis. Consent from the participants were waived as all data was obtained anonymously and it did not include any personal identi cation information. The study protocol was approved by the Ethical Committee for Human Research at the National Cheng Kung University Hospital institutional review board, Taiwan (IRB number: B-ER-108-131).
The baseline data included medical history, medication, smoking status, alcohol use, exercise habit, and tea consumption. Cigarette smoking was calculated in pack-years by multiplying the number of packs of cigarettes smoked per day by the number of years the person has smoked. Subjects who smoked more than one pack of cigarettes per day for half a year were de ned as currently smoking. Exercising for at least three times a week was de ned as regular exercise. The questions for tea consumption included (1) Have you drunk tea habitually once a week for at least six months? (2) How much (in milliliters) tea do you drink each day? (3) How many times do you drink tea in one week? (4) How many years have you been drinking tea in this way? Different examples of tea containers were provided to assist participants to recall the amount of tea consumption. For example, there were traditional Chinese teapot (120 mL), mug (250 mL), aluminum foil packed tea (300 mL), canned tea (350 mL) and bottled tea (600ml). Tea consumption of 120mL per day or more for at least 1 year was de ned as a habitual tea drinker [20]. To obtain the average of tea consumption per day, we multiplied the amount of tea consumed per day and the frequency per week according to the records in the questionnaire. The product was divided by seven to yield the nal result.
Overweight was de ned as body mass index (BMI) of 24 -26.9 kg/ m2 and obesity was de ned as BMI ≥ 27 kg/m2 based on the de nition by the Department of Health in Taiwan [21]. Hypertension was de ned as having a documented history of hypertension, records of currently using anti-hypertensive agents or blood pressure ≥ 140/90 mmHg measured with an automatic blood pressure monitor in a supine position after resting at least ve minutes. Cerebrovascular disease was de ned by a documented history of cerebrovascular disease or transient ischemic attack.
All subjects had blood tests after 12 hours of fasting for fasting plasma glucose, hemoglobin A1c, creatinine, total cholesterol (TC), high density lipoprotein-cholesterol (HDL-C), triglyceride (TG) and C reactive protein (CRP) level. Except for those who were pregnant or had established diagnosis of diabetes on medicine control, all participants were asked to take 75-g glucose and checked for 2-h post-load glucose level. Diabetes mellitus was de ned by a documented history of diabetes, records of currently using anti-diabetic agents, fasting plasma glucose ≥ 126 mg/dl, 2-hour post-load glucose ≥ 200 mg/dl, or hemoglobin A1c ≥ 6.5%. The estimated glomerular ltration rate (eGFR) was calculated by MDRD formula: [22]. The TC/HDL-C ratio was calculated and categorized as either <5.0 or ≥5.0. The CRP levels were strati ed as either one in the highest quartiles or the lower three quartiles with a cutoff of 2.77 mg/dL.

Pulmonary Function Test
The pulmonary function tests were performed by an experienced technician using an automated owsensing spirometer (Multi-Functional Spirometer HI-801; CHEST M.I., Inc) as recommended by the American Thoracic Society's. All participants taking the exam were in a sitting position with nose clip in place. At least two error-free reproducible maneuvers (FEV1 and FVC within 5%) out of three acceptable maneuvers were performed. The participants would need to repeat the maneuvers up to six times if necessary. The parameters obtained in the study were FVC, FEV1, and FEV1 to FVC ratio (FEV1/FVC). RSP was de ned as of FVC < 80% of predicted value and FEV1/FVC ratio ≥ 70% according to the de nition of the American Thoracic Society [23].
Statistical analyses SPSS software (17th version, Chicago, Illinois, USA) was used to analyze statistical data. Continuous variables were presented as the mean ± standard deviation. Categorical variables were presented as numbers (percentages). Independent sample t-test or Mann-Whitney U test was used to compare continuous clinical characteristics, and the Chi-square test was used for categorical parameter comparison between the group with and without RSP. Potential associated confounders for RSP included age, gender, BMI, hypertension, diabetes, cerebrovascular disease, TC/HDL-C ratio, CRP, and regular exercise and they were adjusted in the multiple regression model. The adjusted odds ratio (OR) and the 95% con dence interval (CI) of tea consumption were calculated for its associated risk of RSP. P value < 0.05 was considered as statistically signi cant throughout the analyses.

Results
A total of 1849 subjects (12.3%) had RSP. Table 1 shows the clinical characteristics of subjects with and without RSP. Subjects with RSP were more likely to be older, female, and had higher BMI. RSP group had a higher prevalence of hypertension, diabetes, cerebrovascular disease, eGFR <60, TC/HDL-C ratio ≥ 5, highest quartile of CRP level, and current smoking status. However, subjects with RSP had a lower prevalence of regular exercise. In subjects with and without RSP, the amount of daily tea consumption was 65 ± 191 and 110 ± 242 mL (p<0.001) and the prevalence of habitual tea drinking was 15.1% and 24.0% (p<0.001), respectively. Table 2 shows the relationship between the clinical variables and RSP in smokers and non-smokers based on multiple logistic regression model. In non-smoking group, after adjusting confounders, habitual tea drinkers had a lower associated risk of RSP (model 1, OR = 0.75, CI 0.63-0.89, p < 0.01). Tea consumption ≥ 3 cup-year was inversely related to RSP (model 2, OR = 0.74, CI 0.61-0.89, p < 0.01), while tea consumption <3 cup-year was not. In the smoking group, there was no signi cant association between tea consumption, presented as habitual drinking and tea consumption ≥ 3 cup-year, and RSP.

Discussion
This study showed that habitual tea drinking and tea consumption of at least 3 cup-years had a lower associated risk of RSP in non-smokers but not in smokers. The association was independent of age, gender, BMI, CRP levels, regular exercise and co-morbidities such as hypertension, diabetes and cerebrovascular disease. One cross sectional study conducted in Korea demonstrated that green tea intake ≥2 times per day was associated with an increase in pulmonary function and a reduced associated risk of COPD [19], but RSP was not studied in the study. Two studies investigated the association of catechin and avonoid, the polyphenols abundant in tea, with pulmonary function [24,25]. One population-based study of young Chilean found that comparing those with the highest quintile of intake of catechin versus the lowest quintile, the former group had a signi cantly higher FVC by 70 mL [24]. The Chilean study also applied Food Frequency Questionnaire (FFQ) from which the avonoid content was estimated, but the main dietary sources of these antioxidants in Chilean are fruits and vegetables. The other study was a multi-centric population-based study in European adults and suggested that intake of total avonoid might be related to a lower risk of spirometric restriction [25]. It applied FFQ and US Department of Agriculture (USDA) Database for the Flavonoid Content of Selected Foods to investigate the intake of avonoid that was habitually consumed in the general population.
Though the consumables investigated in these two studies were not tea, considering the effects of the same key components in the tea, the results of the present study lined with those of these studies. In another study examining the relationship between lung function and dietary antioxidants in 680 middleaged European adults over a 10-year period, intake of apples, bananas, tomatoes, herbal tea, and vitamin C was found to be associated with a slower decline in FVC [26]. Unlike the herbal teas commonly consumed in Europe, green, oolong and black teas are the main types of teas consumed in the East. In addition, we further used the term "cup-year" to assess the cumulative effect of tea consumption over time, as in the case of "pack-year" for smoking. We found that tea consumption of ≥3 cup-year was associated with a 26% lower risk of RSP disease in the non-smoking group, but not in the <3 cup-year group. So far as we know, this is the rst study examining the effects of tea consumption on RSP among subjects with different smoking status.
Little is known about the exact mechanisms underlying the association of tea consumption with RSP.
Previous studies showed associations of RSP with heart failure [27], arterial stiffness [28], metabolic syndrome [11,12] and chronic kidney disease [29,30]. The common entity within these diseases or lifestyles are the raised oxidative stress, systemic in ammation and risks of cardiovascular diseases. The antioxidant phenolic compounds, the avonoids, in tea may play a major role in the mechanism. Flavonoids exist in most plant foods, but the concentration is particularly high in tea. The three major classes of avonoids are avonols, avones and catechins. Studies either in vitro or ex vivo have shown that tea extracts, such as avonoids and its secondary metabolites, may participate in manipulating the smooth muscle relaxation, nitric oxide synthase activity in endothelial cells, vascular in ammation reduction, and renin activity inhibition through its anti-in ammatory and anti-oxidative effects [17,31]. In addition, studies also showed that avonoids and tea catechins can signi cantly improve endothelial function [32,33]. Through reducing multiple aspects of cardiovascular risks by improving endothelial system, increasing nitric oxide production, and vascular relaxation, long-term tea consumption may have protective effects on RSP and thereby further reducing lung tissue damages in non-smokers.
In this study, tea consumption was not associated with reduced risks of RSP in the smoking group no matter what cumulative dosage of tea the participant was consuming. (data not shown). One previous study found that regular green tea drinking might protect smokers from DNA damages by eliminating free radicals associated with smoking [34]. Chan, K.H., et al also found green tea might ameliorate the derangement between local oxidative stress and protease/anti-protease in the airways after exposure to cigarette smoking [35]. One potential explanation for the associations was that the injury in the lungs from smoking could be too overwhelming to be compensated by the anti-oxidative effects of tea. This study provided a direction for further studies on the association between tea consumption and RSP in the future.
In this study, aging, female gender, obesity, diabetes, hypertension, highest quartile group of CRP, and less regular exercise were independently related to RSP. The association of these covariates mentioned above and RSP were in agreement with previous studies [6,[8][9][10]36]. As for the relationship between RSP and chronic kidney disease, RSP was reported to be common in patients with advanced chronic kidney disease and the eGFR displayed an inverted J-shaped association with FVC [29,30]. However, this study did not show a signi cantly positive association between chronic kidney disease and RSP. The possible explanation may be related to that most subjects did not have advanced chronic kidney disease (2.5% in non-smokers and 2.2% in smokers). As for lipid pro les, we found TC/HDL-C ≥ 5 had no association with RSP in the adjusted model. A collinearity with diabetes, hypertension, and BMI may result in an insigni cant association of TC/HDL-C≥ 5 with RSP. Another reason may be related to that subjects with dyslipidemia may take anti-hyperlipidemic medications.
This study had several limitations. This is a cross-sectional study. The temporal relationship of tea consumption and the risk of RSP should be generated with caution. The participants enrolled are mainly from southern Taiwan. A more generalized population for further studies are needed. The questionnaires used to determine the amount and frequency of tea consumption depended on the participants' memory, and thus, the recall bias regarding the details of tea intake could not be excluded. Besides, we did not examine the effect of tea by its types. The types of tea in the questionnaires from the health checkup were green tea, oolong tea, black tea, and others. Oolong tea consumers composed the largest population among all tea drinkers. However, the types or brands of tea are miscellaneous in Taiwan and the drinking habits may change overtime. Therefore, we combined the results from all kinds of tea for this study.

Conclusion
In summary, habitual tea drinking and tea consumption of at least 3 cup-years had a reduced odds of RSP in non-smokers but not in smokers. More studies are needed to elucidate the protective effect of tea or its extracts on RSP in both smokers and non-smokers.