Coffee consumption and risk of hypertension in the Polish arm of the HAPIEE cohort study

Abstract

Coffee consumption has been hypothesized to be associated with blood pressure (BP), but previous findings are not homogeneous. The aim of this study was to evaluate the association between coffee consumption and the risk of developing hypertension.
Data on coffee consumption, BP and use of anti-hypertensive medicament were derived from 2725 participants of the Polish arm of the HAPIEE project (Health, Alcohol and Psychosocial factors In Eastern Europe) who were free of hypertension at baseline and followed up for an average of 5 years. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated by multivariate logistic regression analyses and stratified for potential confounding factors.
Coffee consumption was related to decreased age, smoking status and total energy intake. Compared with persons who drink <1 cup coffee per day, systolic BP was significantly associated with coffee consumption and the risk of hypertension was lower for individuals consuming 3-4 cups per day. Despite the analysis stratified by gender showed that the protective effect of coffee consumption on hypertension was significant only in women, the analysis after stratification by smoking status revealed a decreased risk of hypertension in non-smokers drinking 3-4 cups of coffee per day in both sexes (OR 0.41, 95% CI: 0.21, 0.79 for men and OR 0.54, 95% CI: 0.29, 0.99 for women). Upper category coffee consumption (>4 cups per day) was not related to significant increased risk of hypertension.
Relation between coffee consumption and incidence of hypertension was related to smoking status. Consumption of 3-4 cups of coffee per day decreased the risk of hypertension in non-smoking men and women only.European Journal of Clinical Nutrition advance online publication, 29 July 2015; doi:10.1038/ejcn.2015.119.

Full text

ORIGINAL ARTICLE
Coffee consumption and risk of hypertension in the Polish arm
of the HAPIEE cohort study
G Grosso
1,2
, U Stepaniak
2
, M Polak
2
, A Micek
2
, R Topor-Madry
2
, D Stefler
3
, K Szafraniec
2
and A Pajak
2
BACKGROUND/OBJECTIVES: Coffee consumption has been hypothesized to be associated with blood pressure (BP), but previous
findings are not homogeneous. The aim of this study was to evaluate the association between coffee consumption and the risk of
developing hypertension.
SUBJECTS/METHODS: Data on coffee consumption, BP and use of anti-hypertensive medicament were derived from 2725
participants of the Polish arm of the HAPIEE project (Health, Alcohol and Psychosocial factors In Eastern Europe) who were free of
hypertension at baseline and followed up for an average of 5 years. Odds ratios (ORs) and 95% confidence intervals (CIs) were
calculated by multivariate logistic regression analyses and stratified for potential confounding factors.
RESULTS: Coffee consumption was related to decreased age, smoking status and total energy intake. Compared with persons who
drink o1 cup coffee per day, systolic BP was significantly associated with coffee consumption and the risk of hypertension was
lower for individuals consuming 3–4 cups per day. Despite the analysis stratified by gender showed that the protective effect of
coffee consumption on hypertension was significant only in women, the analysis after stratification by smoking status revealed a
decreased risk of hypertension in non-smokers drinking 3–4 cups of coffee per day in both sexes (OR 0.41, 95% CI: 0.21, 0.79 for
men and OR 0.54, 95% CI: 0.29, 0.99 for women). Upper category coffee consumption (44 cups per day) was not related to
significant increased risk of hypertension.
CONCLUSIONS: Relation between coffee consumption and incidence of hypertension was related to smoking status. Consumption
of 3–4 cups of coffee per day decreased the risk of hypertension in non-smoking men and women only.
European Journal of Clinical Nutrition advance online publication, 29 July 2015; doi:10.1038/ejcn.2015.119
INTRODUCTION
Hypertension is one of the most common and important health
problems in the modern world.
1
Elevated blood pressure (BP) is an
established risk factor for coronary artery disease, stroke, kidney
disease, all-cause mortality and decreased life expectancy. The
importance of preventing hypertension by adopting a healthy
lifestyle is undoubted. Together with low physical activity, excess
body weight and unhealthy dietary habits (that is, excessive
sodium and low potassium intakes), coffee consumption has been
considered detrimental owing to results of studies reporting an
association with increased risk of hypertension.
2
However, recent
evidence demonstrated that underrated variables (such as socio-
economic status and concurrent smoking habits) have influenced
the results of previous studies and coffee may, in fact, be
protective against hypertension.
3
Coffee is a mixture of several
compounds that actively influence human homeostasis and
metabolism, including caffeine, phenolic compounds, niacin,
minerals (magnesium, potassium) and fiber. The major beneficial
effects of coffee seems to depend on its content of chlorogenic
acids, a family of polyphenols (mostly caffeic and ferulic acid) that
demonstrates strong antioxidant properties through inhibiting
production of inflammatory mediators.
4
Coffee has been reported
to be the major source of chlorogenic acids in the diet.
5
Although
the acute effect of caffeine intake is to increase BP by blocking
adenosine receptors in the vascular tissue, which leads to
vasoconstriction in the general and micro-circulation,
6
the health
effect of long-term habitual consumption is not clear.
Studying the association between coffee consumption and BP is
challenging because the effects of coffee intake may vary when
considered in short- or long-term, as well as in occasional or
habitual consumers and because of methodological issues.
3,7
Previous cross-sectional studies reported contrasting results,
8,9
but
they may have suffered from reverse causation bias, as people
with high BP could have been advised to decrease coffee
consumption in the past, which would result in a selective
reduction of coffee consumption in hypertensive individuals.
10
Prospective epidemiological studies provide a better methodolo-
gical approach to identify significant factors that may exert an
effect on health during a reasonable long period of time, but they
are scarce and their results are contrasting.
11–13
Also, studies on
urinary caffeine metabolite excretion related to coffee consump-
tion demonstrated certain associations.
14
Several experimental
studies have been conducted demonstrating an acute increase of
BP after consuming coffee (or caffeine), with major limitations of
short duration and of testing relatively high doses of coffee in the
treatment groups.
15
A recent meta-analysis including prospective
observational studies reported inconsistent results,
16
whereas
another pooled analysis including both epidemiological and
experimental studies conducted on hypertensive subjects
reported that caffeine intake produces an acute increase in
BP for ⩾3 h, but there is no association between long-term
coffee consumption and BP.
17
Caffeine, whether considered first
1
Integrated Cancer Registry of Catania-Messina-Siracusa-Enna, Catania, Italy;
2
Department of Epidemiology and Population Studies, Jagiellonian University Medical College,
Krakow, Poland and
3
Department of Epidemiology and Public Health, University College of London, London, UK. Correspondence: Dr G Grosso, Integrated Cancer Registry of
Catania-Messina-Siracusa-Enna, Via Santa Sofia 82, 95123 Catania, Italy.
E-mail: giuseppe.grosso@studium.unict.it
Received 17 October 2014; revised 8 May 2015; accepted 11 June 2015
European Journal of Clinical Nutrition (2015), 1–7
© 2015 Macmillan Publishers Limited All rights reserved 0954-3007/15
www.nature.com/ejcn

responsible for BP alteration, is metabolized by the liver CYP1A2
enzyme, whose activity is in turn influenced by several factors,
such as cigarette smoking and oral contraceptive use.
18
Previous
studies demonstrated that stratification of coffee drinkers and
non-drinkers by smoking habits better describe the intercorrela-
tions between such lifestyle behaviors and BP.
14,19
We previously reported a lower prevalence of metabolic
syndrome in high coffee consumers among Polish participants
of the Health, Alcohol and Psychosocial factors In Eastern Europe
(HAPIEE) study.
9
However, a prospective evaluation over time of
those individuals free of hypertension at baseline was lacking. The
aim of this study was to assess the association between coffee
consumption and the risk of developing hypertension in the
Polish urban sample of the HAPIEE study, also evaluating the
relation between coffee consumption and BP levels assessed at
follow-up visits.
MATERIALS AND METHODS
Study population
The HAPIEE study is a multicenter prospective cohort study investigating
the role of biological, dietary, lifestyle and environmental factors in
cardiovascular and other chronic diseases.
20
The criteria for sample
selection and the methods used were reported in detail elsewhere.
20
Briefly, a random sample of 10 729 subjects (aged 45–69 years) was
recruited at the baseline survey conducted in 2002–2005 (response ratio of
59%) selected from population registers in the urban area of Krakow,
Poland. The participants, after providing written informed consent,
completed a structured questionnaire and were invited for clinical
examination. For the purpose of this analysis, we selected individuals free
of hypertension at baseline and used the following exclusion criteria: (i) did
not agree to undergo follow-up visit; (ii) lack of information on BP levels;
and (iii) lack of answer in more than 50% of food frequency questionnaire.
Application of the selection strategy showed in Figure 1 resulted in the
final selection of 2725 individuals free of hypertension at baseline and
followed for an average of 5 years (last follow-up visit was conducted in
2008). The sample included in this analysis did not substantially differ from
the original sample in background characteristics, but variables of interest,
such as mean BP levels (different by inclusion/exclusion criteria), body
mass index (lower among included individuals), smoking status and coffee
consumption (higher among included individuals) differed as a result of
inclusion/exclusion criteria (Supplementary Additional Table 1).
Dietary assessment and categories of exposure
Dietary data were collected by using a food frequency questionnaire based
on the tool developed by Willett et al.
21
and subsequently adapted in the
Whitehall II Study.
22
The food frequency questionnaires consisted of 148
food and drink items representative of the diet during the preceding
3 months. A country-specific instruction manual that included photo-
graphs to facilitate the estimation of portion sizes was used. Participants
were asked how often, on average, they had consumed that amount of the
item during the last 3 months, with nine responses ranging from ‘never or
less than once per month’to ‘six or more times per day’. Moreover,
participants were asked to include additional foods and frequency of
consumption by manual entry.
Energy and micro-nutrient intakes were calculated through food
composition tables to evaluate sodium and potassium intake.
23
The
average coffee consumption was calculated (in ml) by following the
portion sizes used in the study and then converted into 24-h intake. We
categorized daily coffee consumption according to the standard cup of
coffee (150 ml) in (i) o1 cup/day, (ii) 1–2 cups/day, (iii) 3–4 cups/day and
(iv) 44 cups/day.
Demographic, lifestyle and clinical measurements
Socio-demographic and lifestyle characteristics included age, gender,
educational and occupational level, smoking and alcohol drinking habits,
medicaments and vitamin supplement use. Educational level was
categorized as (i) low (primary/secondary), (ii) medium (high school) and
(ii) high (university). Occupational level was categorized as (i) low
(unskilled/unemployed workers), (ii) medium (partially skilled workers)
and (iii) high (skilled workers). Physical activity included energy
expenditure in leisure time by reporting type and duration of activity
according to the predetermined questionnaire items. The overall amount
of energy expenditure was estimated in kcal/day and categorized in low,
moderate and high activity level. Individuals were categorized according
their smoking status as non-smoker and current smoker. Average alcohol
consumption was categorized as (i) none or low (⩽12 g/day) and (ii)
alcohol drinker (412 g/day). Medicaments use included oral contraceptive,
diabetes and hypercholesterolemia treatment.
Physical examination included measurement of height, weight, waist
circumference and BP using standard procedures.
20
Body mass index was
calculated according to the formula weight (kg)/height(m)
2
. BP was
measured three times at the end of the physical examination and the final
value was the mean among the three measurements. Participants were
considered to have hypertension at baseline visit if they had a systolic and/
or a diastolic BP higher than 139 and 89 mmHg, respectively,
24
or been
taking hypertensive medication within the last 2 weeks. These individuals
were not included in this study. Among those free of hypertension, cases
of new onset hypertension were identified according the aforementioned
criteria evaluated at follow-up visits.
Statistical analysis
Characteristics of the study cohort were described by baseline coffee
consumption categories. Taking into account the natural differences in
hypertension risk between men and women, gender-specific analyses
were conducted. Descriptive presentation relied on cross tabulations.
Continuous variables are presented as means and standard deviations
(s.d.), and categorical variables as frequencies and percentages. Variables
were examined for normality (Kolmogorov). The Chi-square test was used
for comparisons of categorical variables, and the Kruskall–Wallis test was
used for continuous variables.
Figure 1. Flow-chart of individuals included for the analysis.
Coffee consumption and risk of hypertension in Polish adults
G Grosso et al
2
European Journal of Clinical Nutrition (2015) 1 –7 © 2015 Macmillan Publishers Limited

Differences in mean systolic and diastolic BP values by different
categories of coffee consumption were tested by using analysis of
covariance and linear contrast analysis because a dose-dependent effect
was assumed. Similarly, mean differences between follow-up and baseline
systolic and diastolic BP values by different categories of coffee
consumption were calculated.
The association between baseline coffee consumption and incident
hypertension was assessed by logistic regression analyses. Crude
(unadjusted), age-adjusted and full-adjusted models were performed for
the whole sample and separately by gender. Odds ratios and 95%
confidence intervals of having hypertension with the lower category of
coffee consumption as reference were calculated. The full-adjusted
multivariate model was adjusted for variables hypothesized to be
associated with hypertension according literature, such as age, gender,
education, occupation, body mass index, alcohol drinking (stratified in
gender-specific cutoffs, such as 0, 0–12, 12–24, 424 ml/day for men and 0,
0–6, 6–12, 412 ml/day for women), smoking status, physical activity level,
past history of cardiovascular diseases, diabetes at baseline, cholesterol
therapy at baseline, use of oral contraceptive at baseline, total energy
intake, vitamin supplement use, and sodium and potassium intake. As an
interaction was found for smoking status (P= 0.035), the analysis was also
performed after stratification of the sample by such a variable. Statistical
significance was accepted at Po0.05. All statistical analyses were
performed with SPSS for Windows 21.0 (SPSS Inc, Chicago, IL, USA).
RESULTS
The baseline characteristics of the study cohort by coffee
consumption categories are shown in Table 1. There were no
substantial differences between men and women in coffee
consumption. Coffee drinkers were characterized by slightly
younger age, higher prevalence of smoking and had higher total
energy intake. A further significant positive trend by coffee
consumption was found with alcohol consumption among men,
whereas in women, drinking less coffee was related to higher
prevalence of diabetes compared with high consumption. Figure 2
shows that the number of cigarettes per day increased by
increasing category of coffee consumption in both men and
women (Po0.001).
Mean systolic and diastolic BP values by coffee consumption
categories are presented in Table 2. Only systolic BP was
significantly different among the category of exposure, but
gender-stratified analysis revealed that such association was
mostly dependent on the relation present in women. No
significant differences were found for diastolic BP. The change
of systolic and diastolic BP at follow-up by categories of coffee
consumption confirmed such findings, with significant higher
increase in systolic BP in lower rather than higher coffee
consumers in both men and women, despite a linear trend being
found only in the latter (Table 3).
Over an average follow-up period of 5 years, 1735 new cases of
hypertension occurred. The association between baseline coffee
consumption and incidence hypertension is shown in Table 4. The
multivariate-adjusted models demonstrated a decreased risk of
hypertension in participants consuming 3–4 cups of coffee
per day (odds ratio 0.75, 95% confidence interval: 0.58, 0.95)
whereas for those with higher consumption, the increase in risk
was not significant. After stratification by gender, the association
remained significant only in women. Because coffee consumption
was significantly associated with smoking status at univariate
analysis, an additional model was built by stratifying by such a
variable (Table 5). As expected, because smoking was strictly
correlated with coffee drinking, the risk of hypertension among
smokers was not associated with coffee consumption significantly,
whereas among non-smokers, the intake of coffee of up to 3–4
cups per day decreased the risk of hypertension both in men and
in women (odds ratio 0.41, 95% confidence interval: 0.21, 0.79 and
odds ratio 0.54, 95% confidence interval: 0.29, 0.99, respectively).
Table 1. Background characteristics by category of coffee consumption in 2725 participants of the HAPIEE cohort free of hypertension at baseline
Men, coffee consumption Pfor
trend
Women, coffee consumption Pfor
trend
o1 cup/day 1–2 cups/day 3–4 cups/day 44 cups/day o1 cup/day 1–2 cups/day 3–4 cups/day 44 cups/day
No of subjects 334 473 319 25 401 674 479 20
Age (years), mean (s.d.) 56.7 (7.1) 56.3 (8.8) 55.5 (6.9) 56.3 (6.8) 0.038 55.8 (6.6) 55.3 (6.5) 54.5 (6.1) 52.8 (7) 0.001
BMI, mean (s.d.) 26.7 (3.4) 26.6 (3.3) 26.5 (3.4) 25.7 (3.8) 0.349 26.1 (4) 26.2 (4.1) 26.6 (4.1) 25.3 (4.4) 0.252
Baseline systolic blood pressure, mean (s.d.) 123.9 (9.6) 124.3 (9.1) 124.1 (8.3) 124.3 (10.2) 0.786 119.1 (10.9) 119.8 (10.4) 119.2 (10.6) 117.5 (13.3) 0.900
Baseline diastolic blood pressure, mean (s.d.) 78.1 (6.3) 78.4 (6.8) 78.4 (6.1) 78.8 (8.2) 0.504 76.6 (7) 77 (6.7) 76.9 (6.7) 74.8 (6.2) 0.855
Current smoker, n(%) 138 (41.6) 194 (41.3) 147 (46.1) 17 (68) 0.047 118 (29.5) 229 (34.1) 173 (36.2) 11 (57.9) 0.010
Oral contraceptive use, n(%) −−−− 12 (3.0) 27 (4.0) 7 (1.5) 1 (5.0) 0.088
Low educational level, n(%) 30 (9) 37 (7.8) 17 (5.3) 0 0.403 43 (10.8) 58 (8.6) 37 (7.7) 2 (10) 0.155
Low occupational level, n(%) 187 (57.2) 297 (64.8) 186 (60) 11 (48.8) 0.628 167 (43.6) 276 (42.4) 200 (42.7) 6 (30) 0.847
Low physical activity level, n(%) 84 (27) 120 (27.2) 85 (27.9) 5 (20.8) 0.606 102 (26.6) 200 (31) 126 (28) 6 (31.6) 0.853
Alcohol intake 412 g/day, n(%) 11 (3.3) 24 (5.1) 21 (6.6) 4 (16) 0.008 10 (2.5) 19 (2.8) 13 (2.7) 0 0.976
Diabetes or diabetic treatment, n(%) 20 (6) 34 (7.2) 17 (5.3) 0 0.434 19 (4.7) 22 (3.3) 11 (2.3) 0 0.031
Hypercholesterolemia treatment, n(%) 25 (7.5) 44 (9.3) 19 (6) 3 (12) 0.709 49 (12.2) 98 (14.5) 51 (10.6) 3 (15) 0.492
Total energy intake (kcal/day), mean (s.d.) 2136.1 (668.7) 2182.7 (626.9) 2314.4 (705.3) 2347.6 (578.3) o0.001 2091.9 (623.5) 2059.2 (572.1) 2153.6 (605.8) 2365.2 (734.4) 0.032
Vitamin supplement use, n(%) 38 (11.4) 77 (16.3) 51 (16) 2 (8) 0.220 98 (24.4) 165 (24.5) 95 (19.8) 3 (15) 0.063
Sodium intake, mean (s.d.) 3436.2 (1324.1) 3514.3 (1167.6) 3598.4 (1209.9) 3577.3 (1164.7) 0.096 3273.3 (1068.4) 3124.2 (994.1) 3343.6 (1102) 3798.9 (1630) 0.071
Potassium intake, mean (s.d.) 3713 (1366.8) 3722.1 (1189.3) 3874.1 (1342.2) 3729.6 (1030.9) 0.151 3637.3 (1362.4) 3653.9 (1231.4) 3688 (1126.1) 4170.7 (1508.6) 0.270
Abbreviation: BMI, body mass index.
Coffee consumption and risk of hypertension in Polish adults
G Grosso et al
3
© 2015 Macmillan Publishers Limited European Journal of Clinical Nutrition (2015) 1 –7

DISCUSSION
In this study, we prospectively evaluated the association of coffee
consumption with BP and with incidence of hypertension in a
well-established cohort in Eastern Europe, taking into account its
daily intake and potential confounding factors. We found that
excessive coffee consumption (44 cups per day) was not
significantly associated with incident cases of hypertension, which
may be associated with other untested variables or the lack of
statistical power owing to low number of individuals in this
category of exposure. Non-smoking participants consuming up to
3–4 cups of coffee per day had decreased risk (by about 50%) of
hypertension. Coffee consumption correlated inversely with BP
in women.
The acute effects of caffeine intake are well-known,
6
but the
effect of chronic coffee consumption on BP is still unclear. Recent
meta-analysis of randomized clinical trials analyzed the effects of
Figure 2. Incidence of hypertension by coffee consumption and smoking status at baseline in 2725 participants of the HAPIEE cohort.
Table 2. Systolic and diastolic blood pressure measures at follow-up by category of coffee consumption in 2725 participants of the HAPIEE cohort
free of hypertension at baseline
Coffee consumption Pfor trend
o1 cup/day 1–2 cups/day 3–4 cups/day 44 cups/day
Systolic blood pressure, mean (s.d.)
Overall 141.3 (20.9) 140.3 (20.6) 138.5 (19.7) 136.9 (16.2) 0.004
Men 141.3 (19.8) 142.3 (20.4) 140.3 (17.9) 138.8 (19.5) 0.427
Women 141.4 (21.9) 138.9 (20.7) 137.3 (20.8) 134.5 (18.1) 0.003
Diastolic blood pressure, mean (s.d.)
Overall 85.5 (11.5) 86 (11.4) 85.7 (10.9) 86.5 (8.5) 0.697
Men 85.9 (11.1) 86.8 (11.2) 86.5 (10.1) 86.7 (7.3) 0.547
Women 85.2 (11.9) 85.4 (11.5) 85.2 (11.4) 86.2 (10) 0.909
Table 3. Mean differences between follow-up and baseline systolic and diastolic blood pressure measures by category of coffee consumption in
2725 participants of the HAPIEE cohort free of hypertension at baseline
Coffee consumption Pfor trend
o1 cup/day 1–2 cups/day 3–4 cups/day 44 cups/day
Systolic blood pressure, mm Hg (mean, s.d.)
Overall 19.2 (22.5) 17.4 (21.3) 17.0 (21.5) 14.8 (14.9) 0.047
Men 16.3 (20.5) 17.4 (21.0) 16.4 (18.0) 13.5 (15.9) 0.859
Women 21.7 (23.9) 17.4 (21.6) 17.4 (21.8) 16.2 (14.0) 0.013
Diastolic blood pressure, mm Hg (mean, s.d.)
Overall 7.5 (12.6) 7.7 (12.4) 7.9 (11.7) 8.5 (8.1) 0.514
Men 7.2 (11.5) 8.1 (12.1) 8.1 (10.9) 6.6 (8.6) 0.502
Women 7.8 (13.4) 7.4 (12.5) 7.8 (12.1) 10.6 (7.1) 0.752
Coffee consumption and risk of hypertension in Polish adults
G Grosso et al
4
European Journal of Clinical Nutrition (2015) 1 –7 © 2015 Macmillan Publishers Limited

coffee and caffeine intake reporting that BP elevations appeared
to be significant only for caffeine but not for coffee intake.
8
This
finding suggests that despite caffeine acutely increases BP, its
effects may be somehow attenuated whether ingested through
coffee. It is likely that other components of coffee may counteract
the negative effect of caffeine. Experimental studies reported that
while intravenous caffeine increased muscle sympathetic activity
and led to rise of BP in both habitual and non-habitual coffee
drinkers, coffee intake increased BP only in non-habitual drinkers,
maybe due to development of tolerance.
25
These findings may
support the hypothesis that habitual coffee drinkers are less likely
to show a BP response after caffeine intake and less average BP
levels than non-drinkers. Moreover, coffee is rich in BP-lowering
minerals (that is, potassium and magnesium) and antioxidant
compounds (polyphenols) that may outweigh the hypertensive
effects of caffeine.
4
From a mechanistic point of view, genetic and
concurrent lifestyle habits (such as smoking status) may influence
the activity of key enzymes, as CYP1A2, able to metabolize
caffeine and, consequently, influence BP abnormalities. Taking
into account such variables may explain much of the gender
differences often occurring in this type of studies and further
research is needed to better clarify such potential correlations.
Overall, coffee ameliorates metabolism-related parameters
under laboratory and experimental conditions.
26
There is evidence
that the main phenolic compounds of coffee can regulate the
cellular processes leading to inflammatory response.
27
Oxidative
stress has an important role in the process that leads to
metabolism impairment and development of chronic conditions
such as hypertension, because a state of subclinical inflammation
is heavily involved in the pathogenesis of such pathological
conditions. Therefore, consumption of coffee may inhibit inflam-
mation and thus reduce the risk of cardiovascular and other
inflammatory diseases. The antioxidant properties of polyphenols
contained in coffee mostly depend on chlorogenic acids.
28
This
Table 4. Univariate- and multivariate-adjusted odds ratios (95% confidence intervals)
a
of hypertension by categories of coffee consumption in 2725
participants of the HAPIEE cohort free of hypertension at baseline, overall and by gender
Coffee consumption
o1 cup/day 1–2 cups/day 3–4 cups/day 44 cups/day
Overall
Hypertension cases, n(%) 493 (67.1) 739 (64.4) 470 (58.9) 33 (73.3)
Unadjusted 1 0.89 (0.73, 1.08) 0.70 (0.57, 0.86) 1.35 (0.68, 2.66)
Age-adjusted 1 0.89 (0.73, 1.07) 0.70 (0.57, 0.86) 1.41 (0.71, 2.78)
Multivariate
a
1 0.86 (0.68, 1.07) 0.75 (0.58, 0.95) 1.58 (0.85, 3.64)
Men
Hypertension cases, n(%) 224 (67.1) 321 (67.9) 198 (62.1) 21 (84)
Unadjusted 1 1.03 (0.77, 1.39) 0.80 (0.58, 1.11) 2.57 (0.86, 7.69)
Age-adjusted 1 1.05 (0.77, 1.41) 0.81 (0.58, 1.12) 2.63 (0.87, 7.91)
Multivariate
a
1 1.04 (0.73, 1.49) 0.92 (0.62, 1.35) 2.42 (0.66, 8.91)
Women
Hypertension cases, n(%) 269 (67.1) 418 (62) 272 (56.8) 12 (60)
Unadjusted 1 0.80 (0.61, 1.0) 0.64 (0.48, 0.85) 0.74 (0.29, 1.84)
Age-adjusted 1 0.79 (0.61, 1.0) 0.64 (0.48, 0.84) 0.72 (0.28, 1.81)
Multivariate
a
1 0.74 (0.55, 1.0) 0.65 (0.47, 0.90) 1.09 (0.36, 3.33)
a
Adjusted for age, gender (except when analyses were stratified by sex), education, occupation, body mass index, alcohol drinking, smoking status, physical
activity level, past history of cardiovascular diseases, diabetes at baseline, cholesterol therapy at baseline, total energy intake, vitamin supplement use, oral
contraceptives use, and sodium and potassium intakes.
Table 5. Adjusted odds ratios (95% confidence intervals)
a
of hypertension in smokers and non-smokers by baseline categories of coffee
consumption in 2725 participants of the HAPIEE cohort free of hypertension at baseline, overall and by gender
Coffee consumption
o1 cup/day 1–2 cups/day 3–4 cups/day 44 cups/day
Smokers
Overall, multivariate
a
1 0.96 (0.73, 1.26) 0.96 (0.73, 1.26) 1.41 (0.41, 4.86)
Men 1 1.48 (0.94, 2.34) 1.51 (0.91, 2.34) 2.32 (0.22, 24.47)
Women 1 0.73 (0.51, 1.04) 0.73 (0.49, 1.07) 1.01 (0.23, 4.50)
Non-smokers
Overall, multivariate
a
1 0.63 (0.41, 0.96) 0.44 (0.28, 0.69) 0.89 (0.31, 2.53)
Men 1 0.52 (0.27, 0.98) 0.41 (0.21, 0.79) 1.52 (0.28, 8.03)
Women 1 0.75 (0.42, 1.34) 0.54 (0.29, 0.99) 0.68 (0.16, 2.88)
a
Adjusted for age, gender (except when analyses were stratified by sex), education, occupation, body mass index, alcohol drinking, physical activity level, past
history of cardiovascular diseases, diabetes at baseline, cholesterol therapy at baseline, total energy intake, vitamin supplement use, oral contraceptives use,
and sodium and potassium intakes.
Coffee consumption and risk of hypertension in Polish adults
G Grosso et al
5
© 2015 Macmillan Publishers Limited European Journal of Clinical Nutrition (2015) 1 –7

particular family of molecules exhibits anti-inflammatory activity in
a concentration-dependent manner through inhibiting production
of inflammatory mediators (TNF-αand IL-6) in human peripheral
blood mononuclear cells.
4
Investigations during the last decade
have demonstrated their potential anti-hypertensive effects and
beneficial role in improving endothelial and vascular function.
29
Two randomized clinical trials conducted on humans demon-
strated a significant lowering effect on BP of purified chlorogenic
acids up to 10/7 mm Hg after 12 weeks of treatment compared
with placebo.
30,31
Chlorogenic acids have been found to reduce
NAD(P)H-dependent superoxide production,
32,33
to attenuate the
proliferation of vascular smooth muscle cells through inhibiting
intracellular superoxide anion generation,
34
and to interact
with the renin-angiotensin aldosterone system by inhibiting
angiotensin-converting enzyme activity demonstrated both
in vitro and in vivo.
35,36
Among the chlorogenic acid metabolites,
ferulic acid seems to have the greatest effect on BP.
37,38
The
administration of ferulic acid greatly increases nitric oxide
bioavailability and enhanced acetylcholine-induced endothelial-
dependent vasodilation.
39
The main concern regarding the effects of coffee intake on BP
depends on the discrepancy between the results obtained with
purified molecules of chlorogenic acids against the lack of a clear
anti-hypertensive effect of coffee consumption in human studies,
the latter reporting controversial results.
40–42
Findings of pro-
spective cohort studies are indeed inconsistent.
12–14
The overall
effect of coffee consumption on the risk of hypertension has been
outputted in a meta-analysis
43
showing an inverse J-shaped risk,
with an increasing risk of hypertension for low to moderate
consumption and a decreased risk for high consumption of coffee.
There are some aspects of the studies included in the meta-
analysis that should be taken into account. Two of them included
individuals with untreated hypertension.
11,15
Of the studies
conducted on healthy individuals,
12–14
adjusting for important
potential confounding factors (that is, sodium or potassium intake)
or stratified analysis by subgroups of individuals with specific
characteristics that are supposed to influence BP (that is, smokers)
were often lacking. We reported opposite findings, suggesting
that moderate consumption of coffee may even be protective and
the effect of excessive chronic consumption is less evident.
A separate analysis by smoking status determined the conclusions
in our study. From a biological point of view, it seems possible
that, even if the tolerance for the caffeine-induced pressure effect
develops in habitual coffee drinkers,
6,44
the effects of excessive
intake of caffeine reduce such tolerance.
45
Nevertheless, despite
coffee being rich in BP-lowering minerals (that is, potassium and
magnesium) and antioxidants (chlorogenic acid, flavonoids and so
on) that may outweigh potential adverse effects of caffeine,
4
these
compounds could counterbalance caffeine’s pressure effect above
a certain level of consumption.
8
It has been hypothesized that
other compounds produced during the roasting process, may be
responsible for neutralizing the anti-hypertensive effects of
chlorogenic acids. Among the most investigated candidates,
hydroxyhydroquinone, a particular fraction of chlorogenic acids
discovered in roasted coffee after temperature treatment,
demonstrated to eliminate the anti-hypertensive effects of
chlorogenic acids
29,46,47
and, inversely, daily consumption of
hydroxyhydroquinone-free coffee led to a decrease of BP in a
marginally dose-dependent manner.
48
The current hypothesis is
that the production of hydroxyhydroquinone-derived superoxide
may neutralize chlorogenic acid-enhanced nitric oxide bioactivity,
therefore concealing the anti-hypertensive effects of chlorogenic
acids.
47,49
Some limitations of this study should be considered when
interpreting results. Despite our study being population-based
and comprised a large number of men and women from a
homogeneous population, the availability of prospective informa-
tion and the restriction of the analysis to participants free of
hypertension at baseline reduced the overall sample. Moreover,
the sample was recruited in an urban area and cannot be
considered nationally representative and the selection process
may have produced a selection bias toward healthier individuals
with healthier behaviors compared with excluded participants.
However, the number of subjects included is comparable with
previous studies and we consider a strength the inclusion of only
healthy subjects and the assessment of hypertension by direct
measurement of BP together with anti-hypertensive therapy use,
because the use of only drug-treated hypertension as an endpoint
is a serious limitation of previous studies, which probably
contributed to the elimination of undiagnosed hypertension.
However, we found an unusual high prevalence and incidence of
hypertension cases with apparently no reason to explain it.
Another limitation was that information on the pattern of coffee
drinking, such as type, time and brewing method, was limited. The
inclusion in the analysis of decaffeinated coffee depended on its
very small consumption in our sample (which did not allow to
perform the analysis separately). Moreover, information regarded
only the baseline investigation and was self-reported. Finally,
owing to the nature of the investigation, coffee consumption may
be strongly associated with other lifestyle factors that we were not
able to identify and analyze (such as genetic information), and
thus, potential confounders may still remain. However, compared
with previous investigations, we included a large variety of well-
known cofactors associated with both coffee consumption and BP.
Finally, the last category of exposure included fewer individuals
then others and this could have affected the statistical power.
Nevertheless, we preferred to consider for analysis this category as
separate owing to a different biological effect with a lower coffee
consumption.
In conclusion, average consumption of 3–4 cups of coffee
per day decreased the risk of hypertension among non-smokers
and coffee intake seemed to be inversely associated with systolic
BP. In the present study, excessive daily coffee consumption did
not increase the risk of hypertension significantly and background
characteristics, such as smoking status, must be carefully
considered when exploring the effects of coffee consumption.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
ACKNOWLEDGEMENTS
The study has been funded by the Wellcome Trust (grants 064947/Z/01/Z and
081081/Z/06/Z), US National Institute on Ageing (grant 1R01 AG23522-01) and the
MacArthur Foundation Initiative on Social Upheaval and Health (award 71208).
REFERENCES
1 Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J. Global burden
of hypertension: analysis of worldwide data. Lancet 2005; 365:217–223.
2 Grigg D. The worlds of tea and coffee: patterns of consumption. GeoJournal 2003;
57:283–294.
3 Higdon JV, Frei B. Coffee and health: a review of recent human research. Crit Rev
Food Sci Nutr 2006; 46: 101–123.
4 Godos J, Pluchinotta FR, Marventano S, Buscemi S, Li Volti G, Galvano F et al.
Coffee components and cardiovascular risk: beneficial and detrimental effects. Int
J Food Sci Nutr 2014; 21:1–12.
5 Grosso G, Stepaniak U, Topor-Mądry R, Szafraniec K, Pająk A. Estimated dietary
intake and major food sources of polyphenols in the Polish arm of the
HAPIEE study. Nutrition 2014; 30: 1398–1403.
6 Umemura T, Ueda K, Nishioka K, Hidaka T, Takemoto H, Nakamura S et al. Effects
of acute administration of caffeine on vascular function. Am J Cardiol 2006; 98:
1538–1541.
7 Caprioli G, Cortese M, Maggi F, Minnetti C, Odello L, Sagratini G et al.
Quantification of caffeine, trigonelline and nicotinic acid in espresso coffee: the
influence of espresso machines and coffee cultivars. Int J Food Sci Nutr 2014; 65:
465–469.
Coffee consumption and risk of hypertension in Polish adults
G Grosso et al
6
European Journal of Clinical Nutrition (2015) 1 –7 © 2015 Macmillan Publishers Limited

8 Grosso G, Marventano S, Galvano F, Pajak A, Mistretta A. Factors associated
with metabolic syndrome in a Mediterranean population: role of caffeinated
beverages. J Epidemiol 2014; 24:327–333.
9 Grosso G, Stepaniak U, Micek A, Topor-Mądry R, Pikhart H, Szafraniec K et al.
Association of daily coffee and tea consumption and metabolic syndrome: results
from the Polish arm of the HAPIEE study. Eur J Nutr; e-pub ahead of print
4 November 2014.
10 Geleijnse JM. Habitual coffee consumption and blood pressure: an epidemiolo-
gical perspective. Vasc Health Risk Manag 2008; 4:963–970.
11 Uiterwaal CS, Verschuren WM, Bueno-de-Mesquita HB, Ocke M, Geleijnse JM,
Boshuizen HC et al. Coffee intake and incidence of hypertension. Am J Clin Nutr
2007; 85: 718–723.
12 Winkelmayer WC, Stampfer MJ, Willett WC, Curhan GC. Habitual caffeine intake
and the risk of hypertension in women. JAMA 2005; 294: 2330–2335.
13 Palatini P, Dorigatti F, Santonastaso M, Cozzio S, Biasion T, Garavelli G et al.
Association between coffee consumption and risk of hypertension. Ann Med 2007;
39: 545–553.
14 Guessous I, Pruijm M, Ponte B, Ackermann D, Ehret G, Ansermot N et al. Asso-
ciations of ambulatory blood pressure with urinary caffeine and caffeine meta-
bolite excretions. Hypertension 2015; 65:691–696.
15 Noordzij M, Uiterwaal CS, Arends LR, Kok FJ, Grobbee DE, Geleijnse JM. Blood
pressure response to chronic intake of coffee and caffeine: a meta-analysis of
randomized controlled trials. J Hypertens 2005; 23:921–928.
16 Steffen M, Kuhle C, Hensrud D, Erwin PJ, Murad MH. The effect of coffee
consumption on blood pressure and the development of hypertension:
a systematic review and meta-analysis. J Hypertens 2012; 30: 2245–2254.
17 Mesas AE, Leon-Munoz LM, Rodriguez-Artalejo F, Lopez-Garcia E. The effect of
coffee on blood pressure and cardiovascular disease in hypertensive individuals:
a systematic review and meta-analysis. Am J Clin Nutr 2011; 94: 1113–1126.
18 Guessous I, Eap CB, Bochud M. Blood pressure in relation to coffee and caffeine
consumption. Curr Hypertens Rep 2014; 16: 468.
19 Guessous I, Dobrinas M, Kutalik Z, Pruijm M, Ehret G, Maillard M et al. Caffeine
intake and CYP1A2 variants associated with high caffeine intake protect non-
smokers from hypertension. Hum Mol Genet 2012; 21: 3283–3292.
20 Peasey A, Bobak M, Kubinova R, Malyutina S, Pajak A, Tamosiunas A et al.
Determinants of cardiovascular disease and other non-communicable diseases in
Central and Eastern Europe: rationale and design of the HAPIEE study. BMC Public
Health 2006; 6: 255.
21 Willett WC, Sampson L, Stampfer MJ, Rosner B, Bain C, Witschi J et al.
Reproducibility and validity of a semiquantitative food frequency questionnai re.
Am J Epidemiol 1985; 122:51–65.
22 Brunner E, Stallone D, Juneja M, Bingham S, Marmot M. Dietary assessment in
Whitehall II: comparison of 7 d diet diary and food-frequency questionnaire and
validity against biomarkers. Br J Nutr 2001; 86: 405–414.
23 Boylan S, Welch A, Pikhart H, Malyutina S, Pajak A, Kubinova R et al. Dietary habits
in three Central and Eastern European countries: the HAPIEE study. BMC Public
Health 2009; 9: 439.
24 Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr. et al.
Seventh report of the Joint National Committee on Prevention, Detection, Eva-
luation, and Treatment of High Blood Pressure. Hypertension 2003; 42: 1206–1252.
25 Corti R, Binggeli C, Sudano I, Spieker L, Hanseler E, Ruschitzka F et al. Coffee
acutely increases sympathetic nerve activity and blood pressure independently of
caffeine content: role of habitual versus nonhabitual drinking. Circulation 2002;
106: 2935–2940.
26 Salomone F, Li Volti G, Vitaglione P, Morisco F, Fogliano V, Zappalà A et al.
Coffee enhances the expression of chaperones and antioxidant proteins in
rats with nonalcoholic fatty liver disease. Transl Res 2013; 163: 593–602.
27 Ranheim T, Halvorsen B. Coffee consumption and human health--beneficial or
detrimental?--Mechanisms for effects of coffee consumption on different risk
factors for cardiovascular disease and type 2 diabetes mellitus. Mol Nutr Food Res
2005; 49: 274–284.
28 Park JB. Isolation and quantification of major chlorogenic acids in three major
instant coffee brands and their potential effects on H2O2-induced mitochondrial
membrane depolarization and apoptosis in PC-12 cells. Food Funct 2013; 4:
1632–1638.
29 Zhao Y, Wang J, Ballevre O, Luo H, Zhang W. Antihypertensive effects and
mechanisms of chlorogenic acids. Hypertens Res 2012; 35:370–374.
30 Kozuma K, Tsuchiya S, Kohori J, Hase T, Tokimitsu I. Antihypertensive effect of
green coffee bean extract on mildly hypertensive subjects. Hypertens Res 2005;
28:711–718.
31 Watanabe T, Arai Y, Mitsui Y, Kusaura T, Okawa W, Kajihara Y et al. The blood
pressure-lowering effect and safety of chlorogenic acid from green coffee bean
extract in essential hypertension. Clin Exp Hypertens 2006; 28:439–449.
32 Suzuki A, Fujii A, Yamamoto N, Yamamoto M, Ohminami H, Kameyama A et al.
Improvement of hypertension and vascular dysfunction by hydroxyhydroquinone-
free coffee in a genetic model of hypertension. FEBS Lett 2006; 580: 2317–2322.
33 Suzuki A, Yamamoto N, Jokura H, Yamamoto M, Fujii A, Tokimitsu I et al.
Chlorogenic acid attenuates hypertension and improves endothelial function in
spontaneously hypertensive rats. J Hypertens 2006; 24: 1065–1073.
34 Li PG, Xu JW, Ikeda K, Kobayakawa A, Kayano Y, Mitani T et al. Caffeic acid inhibits
vascular smooth muscle cell proliferation induced by angiotensin II in stroke-
prone spontaneously hypertensive rats. Hypertens Res 2005; 28:369–377.
35 Actis-Goretta L, Ottaviani JI, Fraga CG. Inhibition of angiotensin converting
enzyme activity by flavanol-rich foods. J Agric Food Chem 2006; 54:229–234.
36 Ardiansyah, Ohsaki Y, Shirakawa H, Koseki T, Komai M. Novel effects of a single
administration of ferulic acid on the regulation of blood pressure and the hepatic
lipid metabolic profile in stroke-prone spontaneously hypertensive rats. J Agric
Food Chem 2008; 56: 2825–2830.
37 Suzuki A, Kagawa D, Fujii A, Ochiai R, Tokimitsu I, Saito I. Short- and long-term
effects of ferulic acid on blood pressure in spontaneously hypertensive rats. Am J
Hypertens 2002; 15: 351–357.
38 Suzuki A, Kagawa D, Ochiai R, Tokimitsu I, Saito I. Green coffee bean extract and
its metabolites have a hypotensive effect in spontaneously hypertensive rats.
Hypertens Res 2002; 25:99–107.
39 Suzuki A, Yamamoto M, Jokura H, Fujii A, Tokimitsu I, Hase T et al. Ferulic acid
restores endothelium-dependent vasodilation in aortas of spontaneously
hypertensive rats. Am J Hypertens 2007; 20:508–513.
40 Buscemi S, Verga S, Batsis JA, Tranchina MR, Belmonte S, Mattina A et al.
Dose-dependent effects of decaffeinated coffee on endothelial function in
healthy subjects. Eur J Clin Nutr 2009; 63: 1200–1205.
41 Buscemi S, Verga S, Batsis JA, Donatelli M, Tranchina MR, Belmonte S et al. Acute
effects of coffee on endothelial function in healthy subjects. Eur J Clin Nutr 2010;
64:483–489.
42 Buscemi S, Batsis JA, Arcoleo G, Verga S. Coffee and endothelial function: a battle
between caffeine and antioxidants? Eur J Clin Nutr 2010; 64:1242–1243.
43 Zhang Z, Hu G, Caballero B, Appel L, Chen L. Habitual coffee consumption and risk
of hypertension: a systematic review and meta-analysis of prospective observa-
tional studies. Am J Clin Nutr 2011; 93: 1212–1219.
44 Shi J, Benowitz NL, Denaro CP, Sheiner LB. Pharmacokinetic-pharmacodynamic
modeling of caffeine: tolerance to pressor effects. Clin Pharmacol Ther 1993; 53:
6–14.
45 Jee SH, He J, Whelton PK, Suh I, Klag MJ. The effect of chronic coffee drinking on
blood pressure: a meta-analysis of controlled clinical trials. Hypertension 1999; 33:
647–652.
46 Suzuki A, Fujii A, Jokura H, Tokimitsu I, Hase T, Saito I. Hydroxyhydroquinone
interferes with the chlorogenic acid-induced restoration of endothelial function in
spontaneously hypertensive rats. Am J Hypertens 2008; 21:23–27.
47 Ochiai R, Chikama A, Kataoka K, Tokimitsu I, Maekawa Y, Ohishi M et al. Effects of
hydroxyhydroquinone-reduced coffee on vasoreactivity and blood pressure.
Hypertens Res 2009; 32: 969–974.
48 Yamaguchi T, Chikama A, Mori K, Watanabe T, Shioya Y, Katsuragi Y et al.
Hydroxyhydroquinone-free coffee: a double-blind, randomized controlled dose-
response study of blood pressure. Nutr Metab Cardiovasc Dis 2008; 18:408–414.
49 Halliwell B, Long LH, Yee TP, Lim S, Kelly R. Establishing biomarkers of oxidative
stress: the measurement of hydrogen peroxide in human urine. Curr Med Chem
2004; 11:1085–1092.
Supplementary Information accompanies this paper on European Journal of Clinical Nutrition website (http://www.nature.com/ejcn)
Coffee consumption and risk of hypertension in Polish adults
G Grosso et al
7
© 2015 Macmillan Publishers Limited European Journal of Clinical Nutrition (2015) 1 –7