Introduction

Cardiovascular disease (CVD) is a leading cause of morbidity, mortality and disability worldwide (1). High blood pressure (BP) is a major modifiable risk factor for coronary artery disease, stroke, congestive heart failure, and renal dysfunction, and is one of the major risk factors for cardiovascular-related mortality, which accounts for 20% to 50% of all deaths (2). Hypertension affects one billion of people worldwide (3) and about 25.2 millions of adult population in the mainland of China (4). The deleterious impacts of elevated BP on cardiovascular systems have been largely attributed to remodeling of the arterial wall and acceleration of atherosclerosis (5, 6). Traditionally, systolic BP (SBP) and diastolic BP (DBP) are the exclusive mechanical factors applied to predict cardiovascular risks in hypertensive and normotensive individuals (7). For example, a five mm Hg reduction in DBP was reported to reduce the risk of stroke and ischemic heart disease by 34% and 21%, respectively (8). Meanwhile, there is a continuous, consistent, and independent association between BP and cardiovascular mortality (9, 10). More importantly, the risk of death for cardiovascular disease increases progressively throughout the range of BP, including the range of pre-hypertensive BP (9).

Tea is the most widely consumed beverage in many countries in addition to water, especially in Asia (11). World tea production has grown steadily and reached to approximate 4.68 million metric tons in 2012 (12). Tea is produced from the leaves of the plant Camellia sinensis (13), and is traditionally classified into three main types according to the degree of fermentation: black tea (fully fermented), oolong tea (semifermented), and green tea (non-fermented) (14). All the three types of tea contain a lot of biologically active compounds. For instance, green tea contains tea polyphenols (mainly catechins), theanine, and caffeine (15). Black and oolong tea include other components such as the aflavins and thearubigins, which are converted from catechins as a result of enzymatic oxidation during fermentation (14). These active compounds that in animal or human studies engender vasodilatation, attenuating protein degradation (16), protect against endothelial dysfunction and cognitive impairment/dementia (17, 18) and have a lot of beneficial properties (e.g. antioxidant (19), antiinflammatory (20), hypolipidemic (20) and neuroprotective (21)).

A plethora of clinical trials have examined tea consumption- BP relations (22-24). However, the results of these trials were inconsistent and the durations were short. The majority of studies were with relatively modest sample sizes and underpowered to detect modest effects on BP. At the same time, most studies did not have BP as a primary outcome (22-24). An inverse association between tea consumption and BP has been reported in cross-sectional epidemiological studies (6, 25-28). Nevertheless, the multivariable analyses considered a wide range of potential dietary and lifestyle characteristics, rarely controlled in other prospective or controlled trial studies. Thus, it still calls for additional research. In this study, we aimed to examine the associations of tea consumption with BP and the presence of hypertension in older Chinese adults from a large community-based health survey, controlling for potential confounding lifestyle and dietary factors.

Methods

Study population

The study participants were from the Weitang Geriatric Diseases study, a community-based survey conducted in the Weitang town located in Suzhou, an urban metropolis in eastern China. Suzhou is a famous green tea production place, where many people have tea drinking habits. Detailed information about the study design, sampling methods, and representativeness has been published elsewhere (29). Briefly, based on official records, 6,030 person aged 60 years or older residing in the town were invited to participate in this study. Before the study, an invitation letter explaining the nature of the study was sent to each family. From August 2014 to February 2015, 5,613 individuals were included in the Weitang Geriatric Diseases study. Among those 5613 people, a total of 4,611 subjects attended the clinic (response rate 82%). Finally, 4579 individuals completed an interviewer-administered questionnaire, underwent BP and anthropometric examinations, and gave a blood sample and were subsequently included in the current analysis.

The Weitang Geriatric Diseases study was conducted following the tenets of the Helsinki Declaration and was approved by the Institutional Review Board of Soochow University. All participants gave written informed consent at the recruitment stage of the study.

Tea consumption (exposures)

The survey collected detailed information on tea consumption. The first question was “Do you usually drink tea?” Subjects who answered “yes” were coded as habitual tea drinkers and completed the following questions: type (i.e. green tea, black or oolong tea), frequency of tea drinking over the past 12 months (≤1time/week, 2-3 times/week, 4-5 times/week, and 6-7 times/week) and duration of tea drinking (≤ 15 years, 16-30 years and >30 years). As only 173 participants drank tea no more than 5 times per week, we combined and defined them as one group “ ≤5times/week”. A habitual green tea, or black or oolong tea drinker was defined as a participant who habitually consumed corresponding type of tea only, as only a small number of participants (n=9) drank both kinds of tea in the past twelve months.

Blood pressure (dependent variables)

SBP and DBP were measured at least 3 times with a minimum 5-minute interval, by well-trained research nurses. Measurements were performed after participants had been seated for at least 5 minutes after questionnaire completion and at least 30 minutes after blood sampling and had refrained from smoking. The average of the last 2 readings was used in the analysis.

Main covariates

The survey also collected information regarding participants’ socio-demographic conditions (i.e. age, sex, education level, working status, and monthly income); lifestyle-related exposures (i.e. smoking, alcohol use, dietary and outdoor activities); health conditions (e.g. history of chronic conditions such as stroke, heart disease) and medication intake. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. Standard laboratory assays were used to assess serum biomarkers including fasting plasma glucose (FPG; in mg/dL), triglycerides (in mg/dL), highdensity lipoprotein (HDL; in mg/dL) cholesterol, low-density lipoprotein (LDL; in mg/dL) cholesterol, total cholesterols (in mg/dL), glutamic-oxalacetic transaminease (AST, in u/l) and alanine aminotransferase (ALT, in u/l).

Health conditions were determined and adopted in study analysis, including obesity, hypertension, hyperlipidemia, diabetes, history of heart disease and history of stroke. Adults were classified as obese (BMI≥30); hypertension was defined as a mean SBP ≥ 140 mm Hg and/or DBP ≥ 90 mm Hg, and/ or the use of antihypertensive medications; hyperlipidemia as a total cholesterol level ≥6.2 mmol/l or use of lipid-lowering drugs; diabetes as fasting blood-glucose of 7.0 mmol/l or more, or use of diabetic medications; history of heart disease and stroke as self-report of physician diagnosis.

Statistical analysis

The characteristics of participants with and without tea drinking habits were compared using the Student t-test for continuous variables and the chi-square test for categorical variables. Multivariate linear regression models were fitted to explore the relationship of tea consumption with the BP controlling for covariates. Three covariate sets were used for each beverage, respectively:

  1. 1)

    Basic covariate set (model 1): adjusted for age, sex, BMI, smoking status (never, current and former smokers) and alcohol drink status (yes and no);

  2. 2)

    The extend model (model 2) was adjusted for basic covariate set + further adjustment for education level (primary education and below vs. secondary schooling and above), living with a spouse (with vs. without), working status (working vs. retired), monthly income (≤1k CNY, 1001-3000 CNY, >3000 CNY), outdoor activities (with vs. without), dietary (normal vs. vegetarian), family history of hypertension (Yes vs. No), history of diabetes (presence vs. absence), hyperlipidemia (presence vs. absence), heart disease (presence vs. absence) and stroke (presence vs. absence).

  3. 3)

    Full covariate set (model 3): extent model +further adjustment for FPG, total cholesterol, creatinine, uric acid, ALT and AST.

Table 1
figure 1

Characteristics according to tea consumption habits (n =4579 subjects)

Full size image

Subjects were grouped as non-habitual and habitual tea drinkers. With respect to the frequency of tea consumption, participants were categorized into three subgroups: 0 time/per week(non-habitual drinkers), 1-5 times/week and ≥5 times/week. We also created two kinds of categorical variables corresponding to type (green tea vs. other tea) and duration [0 (non-habitual drinkers), 1-15, 15-30, >30 years] of tea consumption. These categorical variables of tea consumption with BP were repeatedly analyzed using multivariate linear models described above. Besides, subgroup analyses were performed, stratifying by BP status (hypertensive, normotensive) and CVD related conditions (CVD risk group was defined if any of hypertension, diabetes, overweight/obese and hyperlipidemia was present). The effect of tea consumption on BP was also checked in those without anti-hypertensive treatment.

In addition, categories of tea intake were entered into a multivariable logistic regression model, with diagnosed hypertension as the dependent variable. We analyzed the independent effects of tea consumption, with stepwise adjustment of covariates in model 1-3 as described before. Each independent variable was expressed with an odds ratio (OR) and 95% confidence interval (95%CI).

Statistical significance was defined as P<0.05 for 2-tailed analysis. All analyses were performed using SAS (version 9.3, 2011, SAS Inc., Cary, NC).

Results

Participant characteristics according to tea consumption

Among 4,579 participants, 3,008 and 1,571 were grouped as non-habitual and habitual tea drinkers, respectively. Demographic, lifestyle, and health factors of participants across two groups were shown in Table 1 and the majority of participants (89.62%) consumed green tea. In addition, the habitual tea drinkers were male-dominant (P<0.001) and more frequently current smokers (P<0.001). Habitual tea drinker was more common among and those who lived without a spouse (P<0.001) or who had outdoor activities (P<0.001). Compared with counterparts, habitual tea drinking was higher among retired person (P<0.001). Prevalence of obesity, hyperlipidemia, diabetes, heart disease, hypertension and stroke were lower among tea consumers, but failed to reach statistical significance.

When examining the association between tea drinking and various anthropometric measures, habitual tea drinkers had better cardiometabolic profiles than non-habitual drinkers, in terms of SBP (P<0.001), total cholesterol (P<0.001), LDL (P<0.001), HDL (P<0.001) and triglycerides (P=0.0003). Meanwhile, the means of creatinine (P<0.001) and uric acid (P<0.001) were significantly higher for participants with tea consumption habits (Table 2).

Table 2
figure 2

Mean and standard deviation of cardiometabolic variables according to tea consumption habits (n =4579 subjects)

Full size image

Association of tea consumption and BP

Table 3 shows the independent linear relationships between tea consumption and BP, controlling for different covariates. A tendency for an inverse relationship of SBP and DBP with tea consumption was observed, indicated by negative β coefficient values for all the models. However, these associations were not consistently statistically significant as described below.

SBP: Compared with non-habitual tea drinker, the habitual tea consumers had significantly lower SBP, although the magnitude of effects varied when adjusted for different covariates (coefficient range: -2.16 to -2.76). The models also revealed that the effects were obvious for higher frequency (coefficient range: -1.18 to -1.52), green tea consumption (coefficient range: -1.73 to -2.06) and longer duration of tea consumption (coefficient range: -0.43 to -0.66).

DBP: Statistically significant and inverse linear dependence of mean DBP on frequency of tea consumption was detected, with the overall negative regression coefficient and inclusion of covariates in adjusted models. Habitual tea drinking, green tea consumption was also associated with significantly lower DBP in full covariate models. Longer duration of tea consumption always had lower DBP, although the difference did not reach significance (coefficient:-0.32 in model 3, P =0.099).

Table 3
figure 3

Regression coefficients relating tea consumption to BP components

Full size image

Multiple linear model analyses, using tea consumption as categorical variables, confirmed our findings in subgroups stratified by status of hypertension. Higher frequency and green tea consumption were associated with lower SBP and DBP in normotensive participants, but there was no such relationship in hypertensive individuals. The frequency of tea intake per week had significant effects on SBP (coefficient:-1.26 in model 3, P =0.027) in those without anti-hypertensive treatment. Meanwhile, we also stratified the subjects by health status into healthy and CVD risk groups (hypertension, diabetes, overweight/obese and hyperlipidemia). The association was more pronounced for CVD healthy participants compared with CVD risk participants.

Association of tea consumption and hypertension

The results of a multiple logistic regression analysis assessing the likelihood of hypertension associated with different categories of tea consumption are given in Table 4. The habit of drinking tea was associated with a lower risk of hypertension [habitual drinker vs. non-habitutal drinker, OR=0.79, 95%CI=0.65-0.95, P=0.011], when we adjusted using covariates in model 3 as we described before. Compared with nonhabitual tea drinkers, the risk of developing hypertension decreased by 16% for those who drank one to five times per week (OR=0.84, 95%CI=0.56-1.27, P=0.408), and was significantly further reduced by 22% for those who drank six times per week or more (OR=0.78, 95%CI=0.64-0.94, P=0.011). When we considered the duration of tea consumption instead of the amount of tea consumption on the risk of hypertension, the risk for hypertension was still significantly lower in those who drank tea for 16-30 years when comparing with the non-tea drinkers. However, increasing the duration of tea consumption for 30 years or more did not offer extra protection against the risk of developing hypertension (the OR for those who had drunk for 16-30 years vs those who had drunk for >30 years was small, as shown in model 3 of Table 4: 0.66 vs 0.85) and there was no significant dose-response relationship between duration and risk of hypertension (P for trend=0.053).

Table 4
figure 4

Multiple logistic regression models of tea consumption of the hypertensive subjects

Full size image

Discussion

In this community-based study of older Chinese adults aged 60 years or older, habitual tea consumption was independently associated with a lower level of BP, even in normotensive population or participants without anti-hypertensive treatment. Nonhabitual tea drinkers were at a higher risk of hypertension than habitual tea drinkers, and there was a progressive reduction in risk associated with higher frequency of tea consumption.

The present study confirmed that adjusted mean SBP decreased with increasing tea consumption, which were consistent with other studies (6, 30).Consistent with other clinical studies (31, 32), the lipid profiles were better in the habitual tea drinkers, compared with non-habitual tea consumer. In the present study, BMI and blood glucose were not different between habitual and non-habitual tea drinkers, similar to other randomized controlled trial (33), which suggested that our study population is not a unique group and that our findings could be extrapolated to other populations.

Because of its high content of catechins, also known as tea flavonoids, tea seems to protect against the development of CVD (7, 34). Catechins could scavenge free radicals, chelate redox active transition-metal ions, inhibit redox active transcription factors, inhibit pro-oxidant enzymes and induce antioxidant enzymes, thereby show their antioxidative properties (31). In addition, catechins showed vasodilator effects, by regulating vascular tone through activating endothelial nitric oxide, inhibiting proliferation of vascular smooth muscle cells through interfering with vascular cell growth factors and preventing vascular inflammation (35-37). Catechins could also inhibit thrombogenesis by suppressing adhesion (32, 38).

As a major risk factor for CVD, hypertension constitutes one of the most common chronic conditions among adults in China (4). The present study confirmed that adjusted mean SBP and risk of hypertension decreased with increasing tea consumption. Although these decreases in BP are modest from an individual’s perspective, such reductions are relevant for cardiovascular health at the population level and may translate to substantial reductions in CVD risk (39). It has been demonstrated that even people who are free of hypertension at the age of 55 years have a 90% lifetime risk for developing hypertension (40). The trend of global aging poses great health and economic challenges and the Chinese population aged 60 and older reached 14.8% by the end of 2013 (41). In addition, accumulating evidence suggests that health-promoting lifestyle modification, which is a cost-effective approach for hypertension control, can improve BP control and even reduce medication needs (42). As one of the most effective measures for preventing hypertension, dietary change has been provide to reduce BP and further have a significant impact on the prevalence of hypertension by previous meta-analysis studies (43, 44). Considering this, our findings may have significant public health implications in the prevention of CVDs, since tea is a very common beverage largely consumed in the whole China.

Some limitations should be considered in light of these results. First, self-reported tea consumption may be subject to misclassification and recall biases, thus may have distorted the effect estimates of tea consumption on BP. Second, similar to most previous studies, information about the way of preparation (filtered, boiled water) or serving size of tea was not available, though the information of frequency was obtained. We did not use a food frequency questionnaire to collect tea consumption data such as serving size, which was due to two reasons. On one hand, the survey questionnaire included a number of questions and rating scales assessing various characteristics of older adults, a balance between the data quantity and quality was therefore inevitable due to respondent fatigue and response burden. On the other hand, unlike other Asian countries (e.g. Japan and Singapore) which normally use a standard volume of tea cup (i.e. 100 mL in Japan (45-48) and about 215 mL in Singapore (49, 50)), teacup and tea were in China vary greatly due to various factors. However, this will be an important distinction to make in future tea-focused dietary trials. Third, the cross-sectional design of the study precludes any conclusions regarding causality. Nevertheless, it adds to the evidence for a possible association between habitual tea intake and BP components and forms the basis for further investigation.

The present study on the relationship between hypertension risk and tea consumption using detailed quantitative and qualitative information with multiple adjusted lifestyle, dietary and cardiometabolic factors. We were aware that a comprehensive link between lifestyle factors and tea consumption and improved control for potential confounders and measurement of the many characteristics of tea, including durations and amounts of consumption in detail. In addition, the effect estimates could be biased in subjects if they were aware of an increased risk of CVD and intentionally changed their diet or lifestyle toward a healthier one. We noted this type of bias in subgroups by excluding subjects with a high blood pressure history and/or those receiving antihypertensive medication.

In conclusion, the present study supports the beneficial effects of tea consumption on BP and hypertension by considering a wide range of potential confounding factors, especially in those elder than 60 years old and especially in participants without hypertension or antihypertensive treatment. The results of this study suggest that habitual tea consumption may help to reduce the disease burden and associated costs of CVDs, subsequently reducing mortality and improving healthrelated quality of life. Randomized control trials are warranted to examine the effectiveness and cost-effectiveness of tea consumption in hypertensive populations to decrease the risk of CVDs, especially in quantifying the volume of tea consumption.

Ethical standards: The Weitang Geriatric Diseases study was conducted comply with the current laws of the People’s Republic of China and following the the experiments comply with the current laws of the country in which they were performed. Tenets of the Helsinki Declaration and was approved by the Institutional Review Board of Soochow University. All participants gave written informed consent at the recruitment stage of the study.

Conflict of Interest Disclosures: No

Author Contributions: Prof. Pan, Dr. Yin and Mr Duan contributed to the study concept and design, methods, data analysis and interpretation of data, and preparation of the manuscript. Mrs Yao, Ms Tu contributed to data acquisition. Dr. Liu and Prof. Xu also contributed to study methods, data analysis, and interpretation of data.

Sponsor’s Role: None.