Pharmacological Research 55 (2007) 187–198

Review

Coffee and cardiovascular disease: In vitro, cellular,

animal, and human studies
Jennifer Stella Bonita, Michael Mandarano, Donna Shuta, Joe Vinson ∗
Department of Chemistry, Loyola Hall, University of Scranton, Scranton, PA 18510, USA
Accepted 19 January 2007

Abstract
Coffee is a commonly consumed beverage with potential health benefits. This review will focus on cardiovascular disease. There are three
preparations of coffee that are commonly consumed and thus worthy of examination; boiled unfiltered coffee, filtered coffee, and decaffeinated
coffee. Coffee has over a thousand chemicals, many formed during the roasting process. From a physiological point of view, the potential
bioactives are caffeine, the diterpenes cafestol and kahweol found in the oil, and the polyphenols, most notably chlorogenic acid. We will
examine coffee and its bioactives and their connection with and effect on the risk factors which are associated with heart disease such as lipids,
blood pressure, inflammation, endothelial function, metabolic syndrome and potentially protective in vivo antioxidant activity. These will be
critically examined by means of in vitro studies, cell experiments, animal supplementation, epidemiology, and the most definitive evidence, human
trials.
© 2007 Elsevier Ltd. All rights reserved.

Keywords: Coffee; Cardiovascular disease; Antioxidant; Epidemiology; Cholesterol; Blood pressure

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
2. In vitro studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
2.1. Diterpenes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
2.2. Caffeine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
2.3. Polyphenols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
3. Cell studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
3.1. Caffeine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
3.2. Polyphenols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
4. Animal studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
4.1. Caffeine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
4.2. Polyphenols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
4.3. Coffee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
5. Human epidemiology studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
5.1. Uric acid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
5.2. Hypertension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
5.3. Inflammation and endothelial function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
5.4. Cardiovascular disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
6. Human supplementation and clinical trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
6.1. Polyphenol absorption and antioxidant activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
6.2. Lipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195

∗ Corresponding author. Tel.: +1 570 941 7551; fax: +1 570 941 7510.
E-mail address: vinson@scranton.edu (J. Vinson).

1043-6618/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phrs.2007.01.006
188
6.3. Blood pressure and vasoreactivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction
Coffee is among the most widely consumed pharmacologi-
cally active beverages in the world. Caffeine is the most widely
consumed psychoactive substance. Since drinking coffee is very
common in Western society, it is important that it be investigated.
For example 52% of all persons in the US over 10 years of
age consume coffee [1]. The latest consumption data for coffee
importing countries from the International Coffee Organization
are from 2004 and found in Fig. 1 [2]. Finland consumes the
most coffee and the United Kingdom the least. The average for
the European Community is 5.1 kg/year and is similar to the US.
Coffee has recently been recommended by a US review panel to
be consumed along with tea in greater quantities than all other
beverages save water. These include caloric beverages such as
milk, non-calorically sweetened beverages, fruit and vegetable
juices, alcohol, sports drinks and calorically sweetened, nutrient-
poor beverages [3]. But is coffee good or bad for the heart? That
is the question that needs to be answered since heart disease is
the number one cause of death in the developed world.
There are two species of coffee trees of commercial impor-
tance, Coffea arabica and Coffea robusta. The two species differ
in chemical composition of the green coffee bean. Arabica con-
tains more lipids and robusta contains more caffeine and sucrose
as well as the polyphenols antioxidant chlorogenic acid and its
derivatives [4]. Due to the fact that arabica has a more desirable
flavor, this variety constitutes 80% of the world trade. Robusta
is often used in instant coffee and as fillers in roast and ground
blends [4]. The roasting process causes a loss of water from the
green bean and degradation of many of the compounds including
the antioxidant polyphenols; however, there is very little differ-
ence in total antioxidants between the different roasts of a bean
[5]. There are three main methods of coffee preparation; boiled
unfiltered coffee, filtered coffee, and decaffeinated coffee, the
latter primarily consumed as instant coffee.
Before examination of the evidence, it is wise to appreci-
ate what are probably the bioactives in coffee. There are over a
thousand compounds, many formed during the roasting process,
which produce the unique taste and smell of coffee [4]. However,
from the point of view of concentration in coffee, prior detec-
tion of the parent compound or metabolites in the body, and
physiological effects, there are essentially only three ingredi-
ents that are important; caffeine, the diterpene alcohols cafestol
and kahweol, and chlorogenic acid and other polyphenols. Their
structures are displayed in Fig. 2. Although caffeine is a major
component of coffee, the caffeine content is highly variable.
A cup of home prepared coffee (150 ml) contains between 30
and 175 mg [6]. In specialty coffees consumed outside the home
the range is 18–80 mg/cup and decaffeinated coffees averaged
5 mg/cup [7]. Coffee is an important source of caffeine; it pro-
vides 71% of the caffeine in the US diet [8]. The diterpenoid
J.S. Bonita et al. / Pharmacological Research 55 (2007) 187–198
195
196
196
alcohols are the oils in coffee and their concentration depends
on the how the coffee is prepared. Filtered coffee has less than
0.1 mg/100 ml, i.e. essentially none, and unfiltered coffee can
have between 0.2 and 18 mg/100 ml depending on the method.
The order of decreasing amounts is the following; Scandinavian
boiled > Turkish/Greek  French press > Espresso  Filter [9].
Boiled coffee has a higher concentration of coffee oils because of
the higher temperature used during its preparation and a longer
contact time between the coffee grounds and water [10,11].
We hypothesize that the bioactive ingredients in coffee relat-
ing to heart disease are the polyphenols which in the body may be
acting as protective antioxidants but also could have other bene-
ficial mechanisms. However, the most well-known ingredient in
coffee is the alkaloid caffeine. This compound has a variety of
pharmacological effects with respect to mood, cognitive perfor-
mance, and motor activity. At present there is no evidence that
caffeine can have any benefit for the heart; on the contrary some
results indicate that it is detrimental in certain conditions. Thus,
there is a Jekyll and Hyde or yin and yang aspect to coffee. We
will attempt to separate the effects of caffeine from the rest of
the compounds in coffee as much as possible.
This work will examine coffee and its ingredients, in par-
ticular the polyphenols, diterpenes, and caffeine, for potential
protective cardiovascular effects on risk factors of heart dis-
ease by means of in vitro experiments, cell and animal studies,
epidemiology, and lastly human trials.
2. In vitro studies
2.1. Diterpenes
The two diterpene compounds cafestol and kahweol are sus-
pected to be the coffee substances which cause the elevation
of serum cholesterol. One mechanism may be the LDL recep-
tor which is involved in the endocytic process of apoB- and
apoE-containing lipoproteins. Cholesterol content of the cell
is regulated via feedback repression of the gene for the LDL
receptor. When the cell is depleted of cholesterol, the LDL
receptor gene is actively transcribed and LDL is removed from
the plasma to replenish the cellular cholesterol. On the other
hand when cholesterol accumulates in the cell, the number of
LDL receptors is down regulated and LDL continues to circu-
late, eventually making its way below the endothelial surface,
becoming oxidized and ultimately producing foam cells and sub-
sequent atherosclerosis [12]. In an elegant in vitro cellular study
Rustan et al. showed the diterpenes reduced the activity of LDL
receptors thus causing extra cellular accumulation of LDL [13].
Although there is a positive side to the coffee oils in that they
have been shown to have anticarcinogenic properties mediated
by induction of phase II enzymes in cell culture studies [14].
 J.S. Bonita et al. / Pharmacological Research 55 (2007) 187–198 189

Fig. 1. Per capita coffee consumption in 2004 for the EU countries, UK, Japan and the US.

2.2. Caffeine 2.3. Polyphenols

Certainly caffeine is a major component in coffee and perhaps
the most often used rationale for drinking it. However, it is our
contention that in order to be beneficial to heart disease caffeine
or any bioactive must improve one of more of the major risk
factors such as lipids or blood pressure or else have a direct or
indirect in vivo antioxidant effect at physiological levels. There
is nothing in the literature to indicate a lipid effect of caffeine
in humans although if it induced a weight loss this could indi-
rectly decrease lipids. Then the question is whether caffeine is
an in vivo antioxidant, in which case it might be bioactive. It has
been shown to be an in vitro scavenger of hydroxyl radicals at
millimolar concentrations using electron spin resonance spec-
troscopy [15]. However, human studies have shown a maximum
plasma concentration of 46 ␮M with four cups of coffee and caf-
feine equivalent to 5 mg/kg of body weight [16]. Caffeine itself
has no LDL antioxidant activity. However, some of the metabo-
lites of caffeine, namely 1-methyxanthine and 1-methyluric acid
are as effective at preventing LDL oxidation as ascorbic acid
at 40 ␮M [17]. Unfortunately these water-soluble demethylated
metabolites have not been detected in plasma using UV detec-
tion although they are seen in urine [18]. This finding is probably
indicative of sub-micromolar plasma levels of the metabolites
which are probably not bioactive although this needs to be
explored further.
The major polyphenol in coffee is chlorogenic acid (CGA)
an ester of caffeic acid and quinic acid as depicted in Fig. 2.
CGA is an in vitro antioxidant with two phenolic groups for
radical scavenging via proton transfer. It has been studied in
a number of models. With two antioxidant assays it was the
poorest antioxidant among seven phenolic acids tested [19]. Cof-
fee has also been extensively studied for antioxidant activity. A
Norwegian group found that coffee had the highest concentra-
tion of polyphenols among the beverages [20]. We used our
Folin–Cocialteu assay with catechin as the standard and coffee
prepared according to the manufacturer’s instructions. Coffee
was then hydrolyzed with base to liberate some bound phenolic
groups. We calculate that the average 180 ml cup of brewed cof-
fee provides 396 mg of polyphenols and instant coffee 316 mg.
Combining this data and US per capita consumption it is appar-
ent that coffee is the number one source of antioxidants in the US
diet [21]. Coffee was also found using other antioxidant assay
methods to be the number one source of antioxidants in Norway
and Spain [22,23].
A literature search indicated that there have been studies of
the effect of CGA on LDL oxidation [24]. Also caffeic acid, a
putative metabolite of CGA, was investigated and found to have
activity [25]. We have compared a number of polyphenols and
flavonoids in an in vitro model of atherosclerosis, namely the

Fig. 2. Chemical structures of proposed bioactive compounds in coffee.

190 J.S. Bonita et al. / Pharmacological Research 55 (2007) 187–198
Fig. 3. Concentration of pure compounds and coffee (antioxidants measured as
catechin equivalents) to inhibit the oxidation of LDL + VLDL 50% compared
with the control.
oxidation of LDL + VLDL, two atherogenic lipoproteins. LDL
is the major carrier of cholesterol and VLDL is the major carrier
of triglycerides. They are isolated from plasma by an affinity col-
umn method and oxidized with cupric ion under physiological
conditions. The polyphenols concentration to inhibit the oxida-
tion 50% (IC50 ) is used for comparison after a dose–response
assay [26]. A comparison of some antioxidant vitamins and other
polyphenols is shown in Fig. 3 [27]. The lower IC50 concentra-
tion indicates stronger antioxidant activity. CGA is moderate in
activity for polyphenols and is much better as an antioxidant
than gallic acid, a black tea component and plasma and urine
metabolite [28]. CGA is much more potent than the vitamin
antioxidants which are poor protectors of lipoprotein oxidation
in this model. The CGA IC50 concentration is certainly physi-
ologically possible but of course the metabolism of CGA must
be taken into account. Surprisingly coffee has the best quality
antioxidants in Fig. 3. Perhaps there is synergism between the
polyphenols in coffee as was found between CGA and catechin
for LDL oxidation [29].
There are two areas where LDL is oxidized, in the plasma and
below the endothelial surface. In the plasma the endogenous pro-
teins, enzymes and antioxidant vitamins inhibit LDL oxidation
effectively. However, oxidation in the sub-endothelial space is
more critical since that is where atherosclerosis begins [30]. It
is our hypothesis that LDL is best protected there by antioxi-
dants which can bind to the LDL and be carried with it below
the endothelium. We have measured LDL + VLDL lipoprotein-
bound antioxidant activity by ex vivo spiking of human plasma
with polyphenols at different concentrations and subsequent
isolation of bound compounds by affinity column chromatog-
raphy. The bound antioxidant activity is then determined by
our standard oxidation protocol and determining the lag time
of conjugated diene oxidation. This time is defined as the inter-
section of the initial slow propagation reaction and the rapid
oxidation reaction when all the antioxidants in the lipoprotein
are oxidized and can no longer protect the lipoproteins from
oxidation [26]. For mixtures such as coffee we determine the
total antioxidants by the Folin–Cocialteu method using cate-
Fig. 4. Plasma spiking with antioxidants followed by isolating LDL + VLDL
and oxidizing under standard conditions to determine the lag time of oxidation
(lipoprotein-bound antioxidant activity) vs. a control.
chin as the standard. Results are shown for ␣-tocopherol, CGA
and brewed coffee in Fig. 4. For comparative purposes the con-
centration to increase the lag time 50% or CLT50 is calculated
from the linear concentration–% lag time increase curve. For
tocopherol it is 50 ␮M, CGA 108 ␮M and coffee 100 ␮M [31].
Tocopherol is the best antioxidant probably due to its highly
lipophilic nature compared with the more water-soluble phenolic
acid CGA and the coffee polyphenols. However, this experiment
demonstrates that the coffee polyphenols can bind to lower den-
sity lipoproteins, protect them from oxidation and thus be in vivo
heart-protective antioxidants.
3. Cell studies
3.1. Caffeine
Cellular studies can be a simple way to determine the
metabolism of coffee compounds and to discover possible mech-
anisms for bioactivity in vivo. Caffeine metabolites are inhibitors
of the enzyme poly(ADP-ribose)polymerase-1 in hydrogen
peroxide-treated epithelial cells at physiological concentrations
[32]. This is a potential in vivo anti-inflammatory caffeine func-
tion in humans and possibly beneficial to the heart.
3.2. Polyphenols
CGA is categorized as a phenolic acid, specifically a hydrox-
ycinnamic acid (see Fig. 2). Metabolism of CGA and other
phenolic acids has been accomplished in human endothelial cells
[33] and most recently in human liver cells [34]. Metabolites
were formed by methylation, sulfation and glucuronidation path-
ways. CGA was found to be part of the most powerful antioxidant
fraction from brewed coffee [35,36]. CGA was effective in pre-
venting oxidative damage to human epithelial cells [37]. Of
interest for heart health is the study with rat cardiomyocytes.
CGA and other hydroxycinnamic acids proved non-cytotoxic,
and they both stabilized membranes and improved the energetic
status of cardiomyocytes [38].
 J.S. Bonita et al. / Pharmacological Research 55 (2007) 187–198
The phenolic acid ferulic acid has been found in coffee [39].
The bioactivity of ferulic acid was investigated after uptake
by cultured human cells and found to be an antioxidant with
equivalent potency to the spice polyphenol curcumin [40]. Pre-
incubation of CGA and other phenolic acids with neuronal
cells showed subsequent protection from peroxy radicals or
hydrogen peroxide by inhibition glutathione depletion, lipid per-
oxidation and reactive oxygen substance formation. [41]. Such
pre-incubation cellular studies demonstrate that CGA and other
polyphenols and/or their metabolites have potential antioxidant
activity in vivo.
Another important cellular study was done by Huang et
al. [42]. Dihydrocaffeic acid (DHCA) is a metabolite of caf-
feic acid (found in coffee) with potent antioxidant properties.
Since DHCA has been detected in human plasma following
coffee ingestion, they tested the hypothesis that DHCA pro-
tects the endothelium from oxidative stress in a model using
human-derived endothelial cells. During culture for 16–24 h,
the cells accumulated DHCA against a concentration gradi-
ent to low millimolar concentrations. In ␣-tocopherol-loaded
cells, DHCA spared ␣-tocopherol during overnight culture in a
dose-dependent manner. In response to oxidant stress induced
by a water-soluble free radical initiator, both ␣-tocopherol and
DHCA diminished oxidation of cis-parinaric acid that had been
incorporated into the cells. This suggests that the protective
effects of DHCA were caused by scavenging of intracellular
reactive oxygen species. DHCA also increased nitric oxide syn-
thase activity in a dose-dependent manner in cultured cells,
which was associated with a comparable increase in endothe-
lial nitric oxide synthase protein (eNOS). This latter mechanism
implies that coffee may prove to protect the heart by improv-
ing endothelial function which is mediated by a decrease in
eNOS. This mechanism has been shown to occur with several
foods and beverages containing polyphenols [43]. Although the
DHCA concentrations required for these effects are higher than
those likely to be present in plasma or the interstitial space,
these results indicate that a coffee metabolite can function as an
intracellular antioxidant.
4. Animal studies
4.1. Caffeine
Animal studies can provide valuable information on the
mechanism of cardiovascular benefit with the proviso that doses
of bioactives and coffee close to that consumed by humans are
used. However, no animal study can with certainty predict the
effect on humans. Chronic caffeine supplementation to preg-
nant monkeys produced no change in cholesterol or triglycerides
which were depressed during pregnancy [44]. In a recent arti-
cle, ovariectomized female rats were investigated as a model
of post-menopausal cholesterol elevation in humans. Catheter-
infused caffeine equivalent to a human dose of two cups of coffee
significantly decreased cholesterol absorption in this animal
model [45].
Although not a heart disease model, the obese diabetic rat
exhibits elevated cholesterol and triglycerides. Caffeine was
191
present at 0.1% in the water which produces plasma caffeine
levels similar to consumption of 300 mg of caffeine for the aver-
age human [46]. This regimen was given for 30 weeks to the
rats. The animals experienced an increase in plasma cholesterol
and an adverse effect on renal function although glucose and
insulin resistance were improved. Caffeine in this model proved
to be harmful for kidney structure and function.
4.2. Polyphenols
The most definitive study of animal CGA metabolites was
done in rats by a French group [47]. Four groups of rats (n = 8)
were fed a diet supplemented with chlorogenic, caffeic or quinic
acids (250 ␮mol/day) or an unsupplemented diet for 8 days. The
recovery of CGA in urine was low (0.8%, mol/mol), and the total
urinary excretion of caffeic acid liberated by hydrolysis of CGA
and its methylated metabolites (ferulic and isoferulic acids) did
not account for >0.5% (mol/mol) of the dose ingested. On the
other hand, the metabolites of microbial origin, namely, m-
coumaric acid and derivatives of phenylpropionic, benzoic and
hippuric acids, represented the major compounds in both urine
and plasma. Hippuric acid largely originated from the transfor-
mation of the quinic acid moiety, and all other metabolites from
the caffeic acid moiety. These microbial metabolites accounted
for 57.4 mole% of the CGA intake. This paper shows that the
bioavailability of CGA depends largely on its metabolism by the
gut microflora in the rat.
A Japanese group investigated the effect of CGA on spon-
taneously hypertensive rats [48]. A single ingestion of CGA
(30–600 mg/kg) reduced blood pressure in spontaneously hyper-
tensive rats, an effect that was blocked by administration of
a nitric oxide synthase inhibitor, N(Gamma)-nitro-l-arginine
methyl ester. The lower dose is still much higher than a common
human dose of 5 mg/kg. When spontaneously hypertensive rats
were fed diets containing 0.5% CGA for 8 weeks (approximately
300 mg/kg per day), the development of hypertension was inhib-
ited compared with the control diet group. Dietary CGA reduced
oxidative stress and improved nitric oxide bioavailability by
inhibiting excessive production of reactive oxygen species in the
vasculature, and led to the attenuation of endothelial dysfunc-
tion. At a more relevant dose, CGA given to mice at 10 mg/kg
activated calcineurin and enhanced macrophage functions in
normal mice, a possible cardiac benefit [49].
Caffeic acid, the initial metabolite of CGA, was given
to the type 2 diabetic mouse (C57BL/KsJ-db/db) to exam-
ine its glucose-lowering and antioxidant effect. Although
this is a diabetic mouse model, hyperglycemia and hyper-
cholesterolemia are oxidative stress conditions. Caffeic acid
significantly increased superoxide dismutase, catalase and glu-
tathione peroxidase and lowered glucose. It also decreased lipid
peroxidation products thus indicating an in vivo antioxidant
activity [50].
4.3. Coffee
The following describes the first animal study on the rela-
tionship of coffee consumption and atherosclerosis [51]. In
192 J.S. Bonita et al. / Pharmacological Research 55 (2007) 187–198
Table 1
Plasma lipids, lipid peroxidation and pooled lag time of LDL + VLDL oxidation (lag time) in hamster given water (control group) or coffee
Group Cholesterol (mM) HDL (mM) LDL + VLDL (mM)
Control 3.72 ± 0.49 0.96 ± 0.16 2.76 ± 0.49
Coffee 3.54 ± 0.96 0.90 ± 0.16 2.64 ± 0.96
Results are mean ± standard deviation.
a Significantly different from control group.
experimental animals the authors investigated the relationship
of coffee consumption with risk factors of atherosclerosis such
as cholesterol, homocysteine, oxidative stress and inflammatory
cytokines. Male Wistar rats were assigned to three treatment
groups (a control diet group, 0.62% coffee diet group, and 1.36%
coffee diet group), and animals were kept on the experimen-
tal diets for 140 days. Coffee diets increased serum caffeine
in a dose–response manner, although caffeine in serum was
not detected in rats fed the control diet. It also led to slightly
increased total serum levels of homocysteine and cholesterol,
but no significant differences were found between the control
and coffee groups. Coffee intake did not affect the production
of IL-6 and TNF-alpha induced by LPS, which contributes to
the atheroma-promoting effect of recurrent bacterial infection.
Biomarkers of oxidative stress, the serum level of 15-isoprostane
F(2t), which was significantly increased by LPS injection, was
not altered by coffee intake. In contrast, urinary 8-hydroxy-2-
deoxyguanosine was significantly increased in the coffee groups
(p < 0.05). From these results, the authors concluded that mod-
erate coffee intake is not a risk factor for atherogenesis.
Our group conducted experiments with hamsters which are a
better model than rats for atherosclerosis. Hamsters, when given
a diet of cholesterol and saturated fat, have a lipid profile similar
to humans. In a short-term study male adult hamsters (five per
group) were given normal rodent food mixed with 0.2% choles-
terol and 10% coconut oil for 2 weeks. The controls were given
water and the experimental group brewed and filtered coffee at
3.6%, i.e. full-strength for humans. Plasma was measured for
cholesterol, HDL, and triglycerides by commercial enzymatic
kits. Lipid peroxides and LDL + VLDL lag time were measured
as described [26]. The coffee group drank significantly less liq-
uid and gained less weight than the control group, p < 0.05.
This has been seen in experiments with caffeine and animals.
Other results are shown in Table 1. There was no effect of cof-
fee on lipids. Lipid peroxidation was significantly decreased
32% and the lag time was almost tripled. Thus coffee is an in
vivo plasma antioxidant and we hypothesize that the polyphe-
nols metabolites from coffee were bound to the LDL + VLDL
and protected it from oxidation similar to our ex vivo studies
previously described in Section 2.3.
A 10-week study was done with the same atherosclerotic
diet as described above. There was a control group and three
experimental groups. For this study commercial instant coffee
at 1/10 the human recipe (described above) was used for the
low decaffeinated and 1/2 the recipe for the high decaffeinated
group. The caffeinated group was given the high dose caffeinated
coffee. There was no significant effect of any dose of coffee,
caffeinated or decaffeinated, on weight gain or food consump-
tion. Also there was no significant effect on lipids due to large
Triglycerides (mM) Lipid peroxidation (␮M) Lag time (min)
0.89 ± 0.51 0.92 ± 0.02 19
1.57 ± 1.40 0.63 ± 0.21a 56
within group variations (data not shown). Although there was
a dose–response decrease in cholesterol with decaffeinated cof-
fee, the caffeinated coffee was equivalent to the decaffeinated
coffee at the high dose. There was no effect of any dose of coffee
on plasma lipid peroxides or atherosclerosis. In confirmation of
our study a French group did not find that CGA or caffeic acid
had any effect on lipids or atherosclerosis in this same hamster
model. These phenolic acids did increase plasma antioxidant
capacity and thus had an in vivo antioxidant effect [52].
5. Human epidemiology studies
Epidemiological evidence can never prove cause and effect
for any regimen such as coffee drinking since it cannot account
for all the variables. For instance coffee drinkers may have a sig-
nificantly less healthy lifestyle than tea drinkers. In a European
study of men women aged 25–64 years, associations of lifestyle
factors with coffee and tea consumption were analyzed [53].
Coffee drinkers smoked more, ate more meat and less fruits than
tea drinkers and were more sedentary. The authors concluded
that drinking coffee is positively associated with factors that
promote coronary heart disease, while drinking tea is associated
with a preventive lifestyle. Also there is a genetic component
which may be significant. It is important to distinguish between
filtered and non-filtered (boiled) coffee since the oils are hyper-
cholesterolemic in humans. Also caffeinated and decaffeinated
coffee may have different physiological effects. Since decaf-
feinated coffee is not popular in Japan and in Europe, only US
studies can properly examine this variable.
5.1. Uric acid
Elevated serum uric acid, although a well-known antioxidant,
is a risk factor for CHD. Uric acid is positively correlated with
aortal intimal media thickness, a measure of atherosclerosis, in
humans [54]. One Japanese epidemiological study is significant
as it is the only one to examine the relationship of coffee drinking
and serum uric acid [55]. There was a clear inverse relationship
between coffee consumption and serum uric acid but none with
green tea, another important source of caffeine in Japan. In a
Polish population of 2000 men and women it was concluded
that coffee drinkers had lower serum uric acid than non-drinkers
and there was a negative correlation between them. Elevated
serum uric acid concentration was positively correlated with
elevated blood pressure and increased body mass index [56].
Clearly further examination of coffee and uric acid needs to be
done by means of supplementation trials.
 J.S. Bonita et al. / Pharmacological Research 55 (2007) 187–198
5.2. Hypertension
Hypertension is a major risk factor for CHD, stroke and
congestive heart failure [57,58]. Any long-term effect on hyper-
tension of caffeine intake in the form of coffee would be of
major news in the healthcare and coffee industries. Caffeine
can raise plasma levels of stress hormones such as epinephrine,
norepinephrine and cortisol, all of which can lead to an increase
in blood pressure [59,60]. The literature on coffee and hyper-
tension has recently been critically reviewed [61]. Interestingly
no association of caffeine with risk of hypertension was found
in the 10 year Nurses Health Study I and II. [62]. Nor were
there any effects of coffee on hypertension. However, there was
an association of consumption of diet and sugared colas and
hypertension, p < 0.001. The risk was 20% higher for >4 cans of
cola/day versus <1 can/day. In a letter to the Editor we attributed
this difference between coffee with no hypertension risk and cola
with risk to the lack of polyphenols in the colas and the presence
of them in the coffee [63]. Thus high consumption of caffeine
with no antioxidants can lead to hypertension at least in women.
5.3. Inflammation and endothelial function
Boiled coffee consumed at moderate to high doses was
also related to increases in C-reactive protein, an inflamma-
tory marker, in a Greek study [64]. Another risk factor for heart
disease is homocysteine. In a Norwegian study 12,000 heavy
coffee-drinking men were examined [65]. Coffee consumption
was found to be a major lifestyle determinant of plasma homo-
cysteine in a healthy population. A similar study with men and
women in Greece reached the same conclusion [66].
The US Nurses Health Study examined coffee drinking and
markers of inflammation and endothelial function in healthy
women and women with type 2 diabetes [67]. In healthy
women, no appreciable differences in plasma concentrations of
the markers were found across categories of caffeinated cof-
fee intake. In women with type 2 diabetes, higher caffeinated
coffee consumption was significantly associated with lower
plasma concentrations of E-selectin and C-reactive protein.
Higher decaffeinated coffee consumption was associated with
lower plasma concentrations of E-selectin and C-reactive pro-
tein only in healthy women. These results indicated that neither
caffeinated nor decaffeinated filtered coffee has a detrimental
effect on endothelial function. On the contrary, the results sug-
gest that filtered coffee consumption is inversely associated with
markers of inflammation and endothelial dysfunction. Interven-
tion studies are needed to definitively conclude the benefits of
coffee in healthy and diabetic men and women.
5.4. Cardiovascular disease
The type of relationship in European studies examining cof-
fee consumption and risk of developing acute coronary events
is a J-curve such as seen in Fig. 5 for a case–controlled study
of a Greek population [68]. The staple coffee is unfiltered and
the mode of preparation is Greek/Turkish (boiled). After con-
trolling for the usual risk factors, moderate coffee drinking
193
Fig. 5. Odds ratio for myocardial infarction (MI) or coronary death relating to
mean coffee consumption in a Greek population of men and women.
(<300 ml/day) reduced the risk 31% relative to no consumption
and higher amounts significantly increased the risk. A simi-
lar J-curve was seen in a Finnish study of middle aged men
[69]. In the latter study there was a dose–response increase in
LDL. Both studies agree that heavy drinking of boiled coffee
(>600 ml/day) is very detrimental to heart health. A Swedish
case–control study elegantly compared boiled unfiltered coffee
with filtered coffee for the risk of the first MI [70]. Men con-
suming >1000 ml/day were over two-fold more likely to have an
MI whether the coffee was filtered or not. Men and women con-
suming boiled coffee, after adjusting for the amount consumed,
had a 40–60% greater risk of MI than those consuming filtered
coffee.
In the largest epidemiological study done to date, long-term
habitual coffee consumption was followed for 16–20 years in
over 44,000 men and 85,000 women free of CHD to assess
heart disease risk in the US Physician’s and Nurses Health
Study [71]. There was no effect of coffee on risk even at >6
cups/day. Regrettably there was no separation of caffeinated
and decaffeinated coffee drinkers. The data does not provide
any evidence that coffee consumption increases the risk of
CHD. The most recent meta-analysis is a compilation of thir-
teen case-control and ten cohort studies followed from 3–44
years [72]. Decaffeinated coffee was not associated with CHD
in four case–control and three cohort studies. There were no
significant differences when considering the regions studied, or
the type of coffee, filtered or unfiltered. The authors conclude
that “despite a significant association between high consump-
tion of coffee and CHD reported among case–control studies,
no significant association between daily coffee consumption
and CHD emerged from long-term follow-up prospective cohort
studies”.
A study of over 40,000 post-menopausal US women was
recently conducted which examined healthy subjects for 15
years [73]. In the fully adjusted model, similar to the rela-
tion of coffee intake to total mortality, the hazard ratio of
death attributed to CVD was 0.76 for consumption of 1–3
cups/day, 0.81 for 4–5 cups/day, and 0.87 for ≥6 cups/day.
194 J.S. Bonita et al. / Pharmacological Research 55 (2007) 187–198
The hazard ratio for death from other inflammatory diseases
was 0.72 for consumption of 1–3 cups/day, 0.67 for 4–5
cups/day, and 0.68 for ≥6 cups/day. Consumption of coffee
may inhibit inflammation and thus reduce the risk of cardio-
vascular and other inflammatory diseases in postmenopausal
women.
Very recently there have been two epidemiological investi-
gations of coffee consumption and myocardial infarction (MI).
In a study of subjects in Costa Rica it was found that the relative
risk of MI in the first hour after coffee drinking was 1.49 [74].
Occasional coffee drinkers (≤1 cup/day) had a very high rela-
tive risk of 4.14 and the risk decreased with increasing coffee
consumption. Heavy drinkers (≥4 cups/day) had a relative risk
of 1.06. When the hazard examination period was extended to
2 or 3 h there was no increase in risk from coffee consumption.
This implicates the acute sympathetic nerve activation of caf-
feine and the development of tolerance to caffeine from heavy
coffee drinking. The Costa Rican population has a very high
intake of saturated fats from tropical oils, an overall low HDL
and an increased risk for MI [75]. The second study was in the
US and investigated almost 2000 subjects who had an acute MI
and survived [76]. There was no association of coffee with post-
infarction mortality, even among the heaviest coffee drinkers.
However, in the first 90 days following an MI there was an
inverse association with mortality from another MI and cof-
fee consumption, i.e. coffee was protective. The authors had
no explanation for this surprising result. It is our hypothesis that
there was a coffee polyphenol-induced improvement in endothe-
lial function of the MI survivors which had been even more
compromised by the initial MI. Also the fact that there was no
relationship between caffeinated cola and mortality rate indi-
cating that caffeine was probably not responsible for the coffee
benefit.
Most people, laypersons and scientists including this author,
believe that consumption of tea is more beneficial than coffee.
However, a recent study in Japan may change our opinion. For
the Japanese, metabolic syndrome is becoming more prevalent
due to the more Western lifestyle adopted by the Japanese, espe-
cially the young. People who have metabolic syndrome are at
considerably elevated risk for CHD [77,78]. The components
of this syndrome are waist circumference, systolic BP (SBP),
diastolic BP (DBP), HDL, triglycerides and fasting blood glu-
cose. In a study of 2000 men and women who were studied for
6 years there was a strong inverse relationship between coffee
consumption and the number of components of the syndrome
for the subjects (see Fig. 6). In contrast, there was no correlation
for green tea drinking [79]. This is certainly paradoxical since
coffee consumption in the West is usually associated with an
unhealthy lifestyle. The effect of coffee on metabolic syndrome
should be examined in randomized controlled clinical trials.
On a lighter note, a Boston group examined the Ben Franklin
hypothesis, namely “early to bed, early to rise, makes a man
healthy, wealthy and wise” and the James Thurber hypothesis,
or “healthy, wealthy and dead”. They found that survivors of MI
were no more likely to die whether they went to bed early or late
or whether they were rich or poor. Nor did sleeping habits have
any relationship with wealth or wisdom (surrogate measure-
Fig. 6. Mean coffee consumption stratified by the number of components of the
metabolic syndrome in Japanese men and women.
ment was education). Not surprisingly hours spent in bed were
significantly and inversely related to cups of coffee consumed
[80].
6. Human supplementation and clinical trials
6.1. Polyphenol absorption and antioxidant activity
A Dutch study examined the human absorption of CGA
and caffeic acid using ileostomy subjects who lack a colon
[81]. Study of this group eliminates bacterial degradation of
the polyphenols which is extensive in normal humans. By urine
analysis they found that CGA and caffeic acid were absorbed 33
and 95% in the body, respectively. Another research team found
that the major metabolites from coffee consumption in normal
human subjects, after hydrolysis of the conjugated metabolites,
were CGA and caffeic acid [82]. An Italian group was the first
to examine polyphenols in human plasma due to coffee. After
drinking 200 ml of coffee, caffeic acid was found in the plasma
and it persisted for 2 h [83]. This same group also found an
increase in plasma antioxidant capacity of 7%, i.e. an in vivo
antioxidant effect, after coffee drinking [84]. This result was
not due to uric acid but to other antioxidants, presumably the
polyphenols from coffee.
In a short-term study with 22 subjects and a control group
it was found that 5 cups/day of unfiltered Italian-style coffee
produced a significant 16% increase in plasma glutathione, a
major in vivo antioxidant. No difference in plasma hydroper-
oxides or homocysteine (both pro-oxidants) was noted in this
intervention study with a coffee intake which was average for
the Italian population [85]. Another study of healthy Finnish men
consuming 3 or 6 cups of filtered coffee for 3 weeks produced
no significant changes in homocysteine, lipid peroxidation or
antioxidant enzymes. Urinary excretion of ferulic and caffeic
acid was increased [86]. A one-week study with healthy male
Japanese subjects given three cups of coffee/day showed that
LDL oxidizability was significantly decreased 30% and returned
to baseline after cessation of coffee [87]. This result corroborates
our ex vivo study and short-term animal study in Sections 2.3
 J.S. Bonita et al. / Pharmacological Research 55 (2007) 187–198
and 4.3 and is indicative of a heart protective mechanism that
should be substantiated in a longer human trial.
6.2. Lipids
Coffee oils from arabica but not robusta raised cholesterol and
triglycerides in 36 normal subjects [88]. Because only arabica
oil has kahweol and arabica coffee contains more cafestol than
does robusta, this implicates the diterpenes as the lipid-raising
ingredients. They were also implicated in the in vitro studies
mentioned in Section 2.1.
The first human trial to convincingly show the different
effects of boiled and filtered coffee on lipids occurred in 1985
[89]. In a 10-week trial 33 hypercholesterolemic men were ran-
domly assigned; (a) to continue with their usual coffee intake;
(b) stop drinking coffee altogether; or (c) stop drinking coffee
for five weeks and thereafter drink either boiled or filter cof-
fee. Cholesterol concentrations fell significantly in all subjects
abstaining for the first 5 weeks compared with subjects con-
suming coffee, and continued to fall in those abstaining for 10
weeks. Cholesterol concentrations rose again in subjects return-
ing to boiled coffee but remained the same in those returning to
filtered coffee. Thus boiled coffee raises cholesterol in humans
compared with filtered coffee.
Some but not all observational studies have demonstrated a
positive association of coffee drinking and higher levels of serum
cholesterol. A total of 14/23 published trials prior to 1989 were
deemed to be well conducted and these were subjected to a meta-
analysis [90]. The authors found that increases in serum lipids
were significantly greater in studies of hyperlipidemic subjects
and in trials of caffeinated or boiled coffee compared with the
few studies with decaffeinated coffee. Trials with filtered coffee
showed very little increase in cholesterol. This is most reassur-
ing for those who drink filtered coffee. The authors concluded
that “consumption of unfiltered, but not filtered, coffee increases
serum levels of total and LDL cholesterol”.
6.3. Blood pressure and vasoreactivity
CGA was examined in human studies for blood pressure and
vasoreactivity effects. There were 28 mild hypertensive sub-
jects who were randomized in a double-blind placebo-controlled
study to receive either 140 mg of CGA (in a green coffee extract)
or placebo for 12 weeks followed by a 2-week washout period
[91]. The CGA regimen had no effect on normal serum or cell
biochemistry. Compared with the placebo group, the CGA group
experienced a significant lowering of SBP and DBP, 10 and
7 mm Hg, respectively. During washout the values increased
to the baseline value in the CGA group. The authors specu-
late that the mechanism of CGA effect is an increase in the
level of nitric oxide (NO). There was no effect on serum magne-
sium by CGA, thus excluding magnesium from being a bioactive
as often mentioned in epidemiological studies of hypertension.
Serum magnesium is inversely related to hypertension risk [92].
Normotensive subjects with reduced vasoreactivity were also
studied with the CGA green coffee bean extract. The same dose
of CGA as in ref 91 was given for 4 months and there was a
195
placebo group [93]. There were mild decreases in DBP and SBP
in the CGA group but they were not significant. There was a
significant decrease in plasma homocysteine compared with the
baseline value for the CGA group. This indicates a decreased
risk of CHD. Also a positive benefit was noted in the vasodila-
tion responses to ischemic reactive hyperemia in the CGA group
indicating improved vasoreactivity, a surrogate for endothelial
function.
The effect of acute administration of caffeine on vascular
function was examined in healthy young men give 300 mg of caf-
feine or a placebo in a double-blind protocol. Blood pressure was
increased by the caffeine but heart rate and forearm blood flow
were not changed. Caffeine was found to improve endothelium-
dependent vasodilation by a mechanism which increased the
production of nitric oxide [94]. In normotensive subjects a sin-
gle dose of caffeine has been shown to cause an increase in
systolic BP of 3–14 mmHg and diastolic BP 4–13 mmHg [95].
A meta-analysis of well conducted clinical trials of coffee or
caffeine before 1997 was undertaken. The median trial duration
was 56 days. Systolic BP increased an average 2.4 and diastolic
BP increased by 1.2 mmHg with coffee compared with control
[96]. They found a greater effect on younger trial subjects. Trials
where the control group consumed no coffee or the small number
that consumed decaffeinated coffee gave the same results. This
indicated that the BP-raising effects are due to caffeine rather
than other ingredients. A more recent meta-analysis from 1966
to 2003 found that regular caffeine consumption significantly
increases BP [97]. When ingested in coffee, however the BP
effect is small. Coffee intake with a median of 725 ml/day pro-
duced only an increase of systolic BP 1.2 mm Hg and diastolic
0.5 mm Hg [95]. In this study caffeine given as tablets produced
BP elevations four times higher than the same dose of caffeinated
coffee. Thus it appears that moderate caffeinated coffee intake
would have no effect on BP.
One of the mechanisms for which coffee or its ingredients
may be detrimental to CV health is mental stress-induced BP
increase. The sympathetic nervous system has an import role in
the regulation of the CV system. Vasomotor sympathetic nerve
activity to skeletal muscle typically increases in response to
mental stress. In patients with borderline hypertension, which
is a large segment of many populations, sympathetic nerve reac-
tivity is already increased under basal conditions and even in
normotensive offspring of hypertensive parents during mental
stress conditions [98]. When given acutely caffeine or coffee, at
a dose which gives the same plasma caffeine, increases sympa-
thetic nerve activity and BP in non-habitual coffee drinkers. On
the other hand, habitual coffee drinkers experienced a lack of BP
increase despite an increased nerve activity from mental stress.
Decaffeinated coffee also increases BP and nerve activity in non-
habitual drinkers. These results point to coffee ingredients other
than caffeine that activate the CV system [99]. It is our contention
that coffee contains substances, presumably polyphenols, able
to reduce the mental stress-related BP increases.
There has been only one long-term human study with decaf-
feinated coffee and BP which lasted 12 weeks [100]. The Dutch
group gave 45 male and female subjects with a habitual intake
of 4–6 cups of coffee per day a regimen of five cups of caf-
196 J.S. Bonita et al. / Pharmacological Research 55 (2007) 187–198
feinated coffee (445 mg dose of caffeine) for 6 weeks or five
cups of decaffeinated coffee (40 mg dose of caffeine) for 6 weeks
then switched to the other form of coffee for 6 weeks. Use
of decaffeinated coffee led to a significant decrease in systolic
and diastolic BP, 1.5 and 1 mmHg, respectively. They conclude
that normotensive subjects switching from caffeinated to decaf-
feinated coffee would experience a small but real decrease.
However as seen in the epidemiology discussion Section 5.2,
consumption of caffeinated coffee does not lead to a greater risk
of hypertension. It is known that chronic alcohol consumption
increases in BP [101]. In fact, 4 weeks of moderate coffee con-
sumption for Japan hypertensive or pre-hypertensive subjects
who also were heavy alcohol drinkers (>60 ml/day) caused a
significant decrease of 7–10 mmHg and 3–7 mmHg in systolic
and diastolic BP, respectively [102]. Thus coffee blunts the BP
increase due to alcohol consumption.
7. Conclusions
Although there have been numerous in vitro studies, cell stud-
ies, animal studies and epidemiology investigations with coffee
and its bioactive components, there has been a paucity of human
trials as shown in this review. This is probably due to the early
evidence that boiled coffee increased lipids in humans. The ini-
tial negative finding was an obstacle to a scientific assessment of
the potential benefits of coffee. From the data presented here, it is
concluded that only heavy consumption (>6 cups/day) of boiled
unfiltered coffee is harmful to the heart as a result of the dose-
related plasma cholesterol and LDL increase due to the diterpene
oils. Although epidemiological studies show that moderate con-
sumption of this coffee appears to confer some cardiovascular
benefit. The CGA effect on blood pressure and endothelial func-
tion is intriguing. Polyphenols are the components in filtered
and unfiltered coffee that have potential cardiovascular benefits
via antioxidant mechanisms related to LDL oxidation as well
as NO bioavailability and blood pressure lowering. However,
their benefit is less obvious when consuming unfiltered coffee.
Polyphenols seem to be countering many of the negative effects
of caffeine and diterpenes in the coffee studies. More long-
term interventional studies need to be done with caffeinated and
decaffeinated coffees in different human populations and disease
states. In the meantime moderate filtered coffee consumption,
which is the usual pattern of the many of the subjects in the
populations studied, is recommended.
References
[1] Barone JJ, Roberts HR. Caffeine consumption. Food Chem Toxicol
1996;34:119–29.
[2] International Coffee Organization, Historical Coffee Statistics, London,
UK 2005.
[3] Popkin BM, Armstrong LE, Bray GM, Caballero B, Frei B, Willett WC.
A new proposed guidance system for beverage consumption in the United
States. Am J Clin Nutr 2006;83:529–42.
[4] Parliament TH, Stahl HB. What makes the coffee smell so good? Chem
Tech 2005:38–47.
[5] Daglia M, Papetti A, Gregotti C, Berte F, Gazzani G. In vitro antioxidant
and ex vivo protective activities of green and roasted coffee. J Agric Food
Chem 2000;48:1449–54.
[6] Gilbert RM, Marshman JA, Schwieder M, Berg R. Caffeine content of
beverages as consumed. Can Med Assoc J 1976;114:205–8.
[7] McCusker RR, Goldberger BA, Cone EJ. Caffeine content of specialty
coffees. J Anal Toxicol 2003;27:520–2.
[8] Frary CD, Johnson RK, Wang MQ. Food sources and intakes of caf-
feine in the diets of persons in the United States. J Am Diet Assoc
2005;105:110–3.
[9] Ranheim T, Halvorsen B. Coffee consumption and human health – ben-
eficial 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;49:274–84.
[10] Aro A, Tuomilehto J, Kostiainen E, Uusitalo U, Pietinen P. Boiled cof-
fee increases serum low density lipoprotein concentration. Metabolism
1987;36:1027–30.
[11] Bonaa K, Arnesen E, Thelle DS, Forde OH. Coffee and cholesterol: is it
all in the brewing? The Tromso Study. BMJ 1988;297:1103–4.
[12] Goldstein JL, Brown MS. Atherosclerosis: the low-density lipoprotein
receptor hypothesis. Metabolism 1977;26:1257–75.
[13] Rustan AC, Halvorsen B, Huggett AC, Ranheim T, Drevon CA.
Effect of coffee lipids (cafestol and kahweol) on regulation of choles-
terol metabolism in HepG2 cells. Arterioscler Thromb Vasc Biol
1997;17:2140–9.
[14] Cavin C, Holzhaeuser D, Scharf G, Constable A, Huber WW, Schilter
B. Cafestol and kahweol, two coffee specific diterpenes with anticarcino-
genic activity. Food Chem Toxicol 2002;40:1155–63.
[15] Shi X, Dalal NS, Jain AC. Antioxidant behaviour of caffeine: efficient
scavenging of hydroxyl radicals. Food Chem Toxicol 1991;29:1–6.
[16] Newton R, Broughton LJ, Lind MJ, Morrison PJ, Rogers HJ, Bradbrook
ID. Plasma and salivary pharmacokinetics of caffeine in man. Eur J Clin
Pharmacol 1981;21:45–52.
[17] Lee C. Antioxidant ability of caffeine and its metabolites based on the
study of oxygen radical absorbing capacity and inhibition of LDL perox-
idation. Clin Chim Acta 2000;295:141–54.
[18] Dobrocky P, Bennett PN, Notarianni LJ. Rapid method for the routine
determination of caffeine and its metabolites by high-performance liquid
chromatography. J Chromatogr 1994;652:104–8.
[19] Yeh CT, Yen GC. Effects of phenolic acids on human phenolsulfo-
transferases in relation to their antioxidant activity. J Agric Food Chem
2003;51:1474–9.
[20] Halvorsen BL, Carlsen MH, Phillips KM, Bohn SK, Holte K, Jacobs
Jr DR, et al. Content of redox-active compounds (i.e. antioxidants) in
foods consumed in the United States. Am J Clin Nutr 2006;84:95–
135.
[21] Vinson JA, Proch J, Bose P, Muchler S, Taffera P, Shuta D, et al. Choco-
late is a powerful ex vivo and in vivo antioxidant, an antiatherosclerotic
agent in an animal model, and a significant contributor to antioxidants
in the European and American diets. J Agric Food Chem 2006;54:
8071–6.
[22] Svilaas A, Sakhi AK, Andersen LF, Svilaas T, Strom EC, Jacobs Jr DR, et
al. Intakes of antioxidants in coffee, wine, and vegetables are correlated
with plasma carotenoids in humans. J Nutr 2004;134:562–7.
[23] Pulido R, Hernandez-Garcia M, Saura-Calixto F. Contribution of bever-
ages to the intake of lipophilic and hydrophilic antioxidants in the Spanish
diet. Eur J Clin Nutr 2003;57:1275–82.
[24] Laranjinha JA, Almeida LM, Madeira VM. Reactivity of dietary
phenolic acids with peroxyl radicals: antioxidant activity upon low
density lipoprotein peroxidation. Biochem Pharmacol 1994;48:487–
94.
[25] Nardini M, D’Aquino M, Tomassi G, Gentili V, Di Felice M, Scac-
cini C. Inhibition of human low-density lipoprotein oxidation by caffeic
acid and other hydroxycinnamic acid derivatives. Free Radic Biol Med
1995;19:541–52.
[26] Vinson JA, Proch J, Bose P. Determination of quantity and quality
of polyphenol antioxidants in foods and beverages. Methods Enzymol
2001;335:103–14.
[27] Vinson JA, Dabbagh YA, Serry MM, Jang J. Plant flavonoids, especially
tea flavonols, are powerful antioxidants using an in vitro oxidation model
for heart disease. J Agric Food Chem 1995;43:2800–2.
 J.S. Bonita et al. / Pharmacological Research 55 (2007) 187–198
[28] Shahrzad S, Aoyagi K, Winter A, Koyama A, Bitsch I. Pharmacokinetics
of gallic acid and its relative bioavailability from tea in healthy humans.
J Nutr 2001;131:1207–10.
[29] Shafiee M, Carbonneau MA, d’Huart JB, Descomps B, Leger CL. Syn-
ergistic antioxidative properties of phenolics from natural origin toward
low-density lipoproteins depend on the oxidation system. J Med Food
2002;5:69–78.
[30] Witztum JL, Steinberg D. The oxidative modification hypothesis of
atherosclerosis: does it hold for humans? Trends Cardiovasc Med
2001;11:93–102.
[31] Vinson JA, Jang J, Yang J, Dabbagh Y, Liang X, Serry M, et al. Vitamins
and especially flavonoids in common beverages are powerful in vitro
Antioxidants which enrich lower density lipoproteins and increase their
oxidative Resistance after ex vivo spiking in human plasma. J Agric Food
Chem 1999;47:2502–4.
[32] Geraets L, Moonen HJ, Wouters EF, Bast A, Hageman GJ.
Caffeine metabolites are inhibitors of the nuclear enzyme poly(ADP-
ribose)polymerase-1 at physiological concentrations. Biochem Pharma-
col 2006;72:902–10.
[33] Kern SM, Bennett RN, Needs PW, Mellon FA, Kroon PA, Garcia-Conesa
MT. Characterization of metabolites of hydroxycinnamates in the in vitro
model of human small intestinal epithelium caco-2 cells. J Agric Food
Chem 2003;51:7884–91.
[34] Mateos R, Goya L, Bravo L. Uptake and metabolism of hydroxycinnamic
acids (chlorogenic, caffeic, and ferulic acids) by HepG2 cells as a model
of the human liver. J Agric Food Chem 2006;54:8724–32.
[35] Daglia M, Racchi M, Papetti A, Lanni C, Govoni S, Gazzani G. In vitro
and ex vivo antihydroxyl radical activity of green and roasted coffee. J
Agric Food Chem 2004;52:1700–4.
[36] Somoza V, Lindenmeier M, Wenzel E, Frank O, Erbersdobler HF, Hoff-
man T. Activity-guided identification of a chemopreventive compound
in coffee beverages using in vitro and in vivo techniques. J Agric Food
Chem 2003;51:6861–9.
[37] Graziani G, D’Argenio G, Tuccillo C, Loguercio C, Ritieni A, Morisco F,
et al. Apple polyphenol extracts prevent damage to human gastric epithe-
lial cells in vitro and to rat gastric mucosa in vivo. Gut 2005;54:193–
200.
[38] Chlopcikova S, Psotova J, Miketova P, Sousek J, Lichnovsky V, Simanek
V. Chemoprotective effect of plant phenolics against anthracycline-
induced toxicity on rat cardiomyocytes. Part II. Caffeic, chlorogenic and
rosmarinic acids. Phytother Res 2004;18:408–13.
[39] Fujioka K, Shibamoto T. Quantitation of volatiles and nonvolatile acids
in an extract from coffee beverages: correlation with antioxidant activity.
J Agric Food Chem 2006;54:6054–8.
[40] Ferguson LR, Zhu ST, Harris PJ. Antioxidant and antigenotoxic effects
of plant cell wall hydroxycinnamic acids in cultured HT-29 cells. Mol
Nutr Food Res 2005;49:585–93.
[41] Pavlica S, Gebhardt R. Protective effects of ellagic and chlorogenic acids
against oxidative stress in PC12 cells. Free Radic Res 2005;39:1377–90.
[42] Huang J, de Paulis T, May JM. Antioxidant effects of dihydrocaffeic acid
in human EA.hy926 endothelial cells. J Nutr Biochem 2004;15:722–9.
[43] Stoclet JC, Chataigneau T, Ndiaye M, Oak MH, El Bedoui J, Chataigneau
M, et al. Vascular protection by dietary polyphenols. Eur J Pharmacol
2004;500:299–313.
[44] Gilbert SG, Rice DC. Effects of chronic caffeine consumption in preg-
nant monkeys (Macaca fascicularis) on blood and urine clinical chemistry
parameters. Fundam Appl Toxicol 1991;16:299–308.
[45] Wang S, Noh SK, Koo SI. Epigallocatechin gallate and caffeine dif-
ferentially inhibit the intestinal absorption of cholesterol and fat in
ovariectomized rats. J Nutr 2006;136:2791–6.
[46] Tofovic SP, Kost Jr CK, Jackson EK, Bastacky SI. Long-term caffeine
consumption exacerbates renal failure in obese, diabetic, ZSF1 (fa-fa(cp))
rats. Kidney Int 2002;61:1433–44.
[47] Gonthier MP, Verny MA, Besson C, Remesy C, Scalbert A. Chloro-
genic acid bioavailability largely depends on its metabolism by the gut
microflora in rats. J Nutr 2003;133:1853–9.
[48] Suzuki A, Yamamoto N, Jokura H, Yamamoto M, Fujii A, Tokim-
itsu I, et al. Chlorogenic acid attenuates hypertension and improves
197
endothelial function in spontaneously hypertensive rats. J Hypertens
2006;24:1065–73.
[49] Wu HZ, Luo J, Yin YX, Wei Q. Effects of chlorogenic acid, an
active compound activating calcineurin, purified from Flos Lonicerae on
macrophage. Acta Pharmacol Sin 2004;25:1685–9.
[50] Jung UJ, Lee MK, Park YB, Jeon SM, Choi MS. Antihyperglycemic and
antioxidant properties of caffeic acid in db/db mice. J Pharmacol Exp
Ther 2006;318:476–83.
[51] Sakamoto W, Isomura H, Fujie K, Takahashi K, Nakao K, Izumi H. Rela-
tionship of coffee consumption with risk factors of atherosclerosis in rats.
Ann Nutr Metab 2005;49:149–54.
[52] Auger C, Laurent N, Laurent C, Besancon P, Caporiccio B, Teissedre
PL, et al. Hydroxycinnamic acids do not prevent aortic atherosclerosis in
hypercholesterolemic golden Syrian hamsters. Life Sci 2004;74:2365–77.
[53] Schwarz B, Bischof HP, Kunze M. Coffee, tea, and lifestyle. Prev Med
1994;23:377–84.
[54] Erdogan D, Gullu H, Caliskan M, Yildirim E, Bilgi M, Ulus T, et al.
Relationship of serum uric acid to measures of endothelial function and
atherosclerosis in healthy adults. Int J Clin Pract 2005;59:1276–82.
[55] Kiyohara C, Kono S, Honjo S, Todoroki I, Sakurai Y, Nishiwaki M, et al.
Inverse association between coffee drinking and serum uric acid concen-
trations in middle-aged Japanese males. Br J Nutr 1999;82:125–30.
[56] Olak-Bialon B, Marcisz C, Jonderko G, Olak Z, Szymszal J, Orzel A.
Does coffee drinking influence serum uric acid concentration? Wiad Lek
2004;57(Suppl. 1):233–7.
[57] Fiebach NH, Hebert PR, Stampfer MJ, Colditz GA, Willett WC, Rosner
B, et al. A prospective study of high blood pressure and cardiovascular
disease in women. Am J Epidemiol 1989;130:646–54.
[58] Vasan RS, Levy D. The role of hypertension in the pathogenesis
of heart failure. A clinical mechanistic overview. Arch Intern Med
1996;156:1789–96.
[59] Robertson D, Frolich JC, Carr RK, Watson JT, Hollifield JW, Shand DG,
et al. Effects of caffeine on plasma rennin activity, catecholamines and
blood pressure. N Engl J Med 1978;298:181–6.
[60] Lane JD, Adcock RA, Williams RB, Kuhn CM. Caffeine effects on cardio-
vascular and neuroendocrine responses to acute psychosocial stress and
their relationship to level of habitual caffeine consumption. Psychosom
Med 1990;52:320–36.
[61] Hamer M. Coffee and health: explaining conflicting results in hyperten-
sion. J Hum Hypertens 2006;20:909–12.
[62] Winkelmayer WC, Stampfer MJ, Willett WC, Curhan GC. Habit-
ual caffeine intake and the risk of hypertension in women. JAMA
2005;294:2330–5.
[63] Vinson JA. Caffeine and incident hypertension in women. JAMA
2006;295:2135.
[64] Zampelas A, Panagiotakos DB, Pitsavos C, Chrysohoou C, Stefanadis C.
Associations between coffee consumption and inflammatory markers in
healthy Persons: the ATTICA study. Am J Clin Nutr 2004;80:862–7.
[65] Refsum H, Nurk E, Smith AD, Ueland PM, Gjesdal CG, Bjelland I, et
al. The Hordaland Homocysteine Study: a community-based study of
homocysteine, its determinants, and associations with disease. J Nutr
2006;136:1731S–40S.
[66] Chrysohoou C, Panagiotakos DB, Pitsavos C, Zeimbekis A, Zampelas
A, Papademetriou L, et al. The associations between smoking, physical
activity, dietary habits and plasma homocysteine levels in cardiovascular
disease-free people: the ‘ATTICA’ study. Vasc Med 2004;9:117–23.
[67] Lopez-Garcia E, van Dam RM, Qi L, Hu FB. Coffee consumption and
markers of inflammation and endothelial dysfunction in healthy and dia-
betic women. Am J Clin Nutr 2006;84:888–93.
[68] Panagiotakos DB, Pitsavos C, Chrysohoou C, Kokkinos P, Toutouzas P,
Stefanadis C. The J-shaped effect of coffee consumption on the risk of
developing acute coronary syndromes: the CARDIO2000 case–control
study. J Nutr 2003;133:3228–32.
[69] Happonen P, Voutilainen S, Salonen JT. Coffee drinking is dose-
dependently related to the risk of acute coronary events in middle-aged
men. J Nutr 2004;134:2381–6.
[70] Hammar N, Andersson T, Aldredsson L, Reuterwall C, Nilsson T, Hal-
lqvist J, et al. SHEEP and the VHEEP study. Association of boiled and
198 J.S. Bonita et al. / Pharmacological Research 55 (2007) 187–198
filtered coffee with incidence of first nonfatal myocardial infarction: the
SHEEP and the VHEEP study. J Intern Med 2003;253:653–9.
[71] Lopez-Garcia E, van Dam RM, Willett WC, Rimm EB, Manson JE,
Stampfer MJ, et al. Coffee consumption and coronary heart disease in men
and women: a prospective cohort study. Circulation 2006;13:2045–53.
[72] Sofi F, Conti AA, Gori AM, Eliana Luisi ML, Casini A, Abbate R, et al.
Coffee consumption and risk of coronary heart disease: a meta-analysis.
Nutr Metab Cardiovasc Dis 2007;17:209–23.
[73] Andersen LF, Jacobs Jr DR, Carlsen MH, Blomhoff R. Consumption of
coffee is associated with reduced risk of death attributed to inflammatory
and cardiovascular diseases in the Iowa Women’s Health Study. Am J
Clin Nutr 2006;83:1039–46.
[74] Baylin A, Hernandez-Diaz S, Kabagambe EK, Siles X, Campos H.
Transient exposure to coffee as a trigger of a first nonfatal myocardial
infarction. Epidemiology 2006;17:506–11.
[75] Martinez-Ortiz JA, Fung TT, Baylin A, Hu FB, Campos H. Dietary pat-
terns and risk of nonfatal acute myocardial infarction in Costa Rican
adults. Eur J Clin Nutr 2006;60:770–7.
[76] Mukamal KJ, Maclure M, Muller JE, Sherwood JB, Mittleman MA.
Caffeinated coffee consumption and mortality after acute myocardial
infarction. Am Heart J 2004;147:999–1004.
[77] Alexander CM, Landsman PB, Teutsch SM, Haffner SM. Third
National Health and Nutrition Examination Survey (NHANES III):
National Cholesterol Education Program (NCEP). NCEP-defined
metabolic syndrome, diabetes, and prevalence of coronary heart dis-
ease among NHANES III participants age 50 years and older. Diabetes
2003;52:1210–4.
[78] Kraja AT, Borecki IB, North K, Tang W, Myers RH, Hopkins PN, et al.
Longitudinal and age trends of metabolic syndrome and its risk factors:
The Family Heart Study. Nutr Metab 2006;3:41.
[79] Hino A, Adachi H, Enomoto M, Furuki K, Shigetoh Y, Ohtsuka M, et
al. Habitual coffee but not green tea consumption is inversely associated
with metabolic syndrome. An epidemiological study in a general Japanese
population. Diabetes Res Clin Pract 2007;76:383–9.
[80] Mukamal KJ, Wellenius GA, Mittleman MA. Holiday review. Early to
bed and early to rise: does it matter? CMAJ 2006;175:1560–2.
[81] Olthof MR, Hollman PC, Katan MB. Chlorogenic acid and caffeic acid
are absorbed in humans. J Nutr 2001;131:66–71.
[82] Mennen LI, Sapinho D, Ito H, Bertrais S, Galan P, Hercberg S, et
al. Urinary flavonoids and phenolic acids as biomarkers of intake for
polyphenol-rich foods. Br J Nutr 2006;96:191–8.
[83] Nardini M, Cirillo E, Natella F, Scaccini C. Absorption of pheno-
lic acids in humans after coffee consumption. J Agric Food Chem
2002;50:5735–41.
[84] Natella F, Nardini M, Giannetti I, Dattilo C, Scaccini C. Coffee drinking
influences plasma antioxidant capacity in humans. J Agric Food Chem
2002;50:6211–6.
[85] Esposito F, Morisco F, Verde V, Ritieni A, Alezio A, Caporaso N, et
al. Moderate coffee consumption increases plasma glutathione but not
homocysteine in healthy subjects. Aliment Pharmacol Ther 2003;17:595–
601.
[86] Mursu J, Voutilainen S, Nurmi T, Alfthan G, Virtanen JK, Rissanen TH,
et al. The effects of coffee consumption on lipid peroxidation and plasma
total homocysteine concentrations: a clinical trial. Free Radic Biol Med
2005;38:527–34.
[87] Yukawa GS, Mune M, Otani H, Tone Y, Liang XM, Iwahashi H, et al.
Effects of coffee consumption on oxidative susceptibility of low-density
lipoproteins and serum lipid levels in humans. Biochemistry (Mosc)
2004;69:70–4.
[88] van Rooij J, van der Stegen GH, Schoemaker RC, Kroon C, Burggraaf J,
Hollaar L, et al. A placebo-controlled parallel study of the effect of two
types of coffee oil on serum lipids and transaminases: identification of
chemical substances involved in the cholesterol-raising effect of coffee.
Am J Nutr 1995;61:1277–83.
[89] Forde OH, Knutsen SF, Arnesen E, Thelle DS. The Tromso heart
study: coffee consumption and serum lipid concentrations in men with
hypercholesterolaemia: a randomised intervention study. Br Med J
1985;290:893–5.
[90] Lee SH, He J, Appel LJ, Whelton PK, Suh I, Klag MJ. Coffee consumption
and serum lipids: a meta-analysis of randomized controlled clinical trials.
Am J Epidemiol 2001;153:353–62.
[91] 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;28:439–49.
[92] Song Y, Sesso HD, Manson JE, Cook NR, Buring JE, Liu S. Dietary
Magnesium Intake and Risk of Incident Hypertension Among Middle-
Aged and Older US Women in a 10-Year Follow-Up Study. Am J Cardiol
2006;98:1616–2162.
[93] Ochiai R, Jokura H, Suzuki A, Tokimutsu I, Ohishi M, Komai N, et al.
Green coffee bean extract improves human vasoreactivity. Hypertens Res
2004;27:731–7.
[94] 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;98:1538–41.
[95] Nurminen ML, Nittynen L, Korpela R, Vapaatalo H. Coffee, caffeine and
blood pressure: a critical review. Eur J Clin Nutr 1999;53:891–9.
[96] Lee 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;33:647–52.
[97] Noordzij M, Uiterwaal CS, Arends LR, Kok FJ, Grobbee DE, Gelei-
jnse JM. Blood pressure response to chronic intake of coffee and
caffeine: a meta-analysis of randomized controlled trials. J Hypertens
2005;23:921–8.
[98] Noll G, Wenzel RR, Schneider M, Oesch V, Binggeli C, Shaw S, et al.
Increased activation of sympathetic nervous system and endothelin by
mental stress in normotensive offspring of hypertensive parents. Circula-
tion 1996;93:866–9.
[99] Sudano I, Spieker L, Binggeli C, Ruschitzka F, Luscher TF, Noll G, et al.
Coffee blunts mental stress-induced blood pressure increase in habitual
but not in nonhabitual coffee drinkers. Hypertension 2005;46:521–6.
[100] van Dusseldorp M, Smits P, Thien T, Katan MB. Effect of decaffeinated
versus regular coffee on blood pressure. A 12-week, double-blind trial.
Hypertension 1989;14:563–9.
[101] Klatsky AL, Koplik S, Gunderson E, Kipp H, Friedman GD. Sequelae
of systemic hypertension in alcohol abstainers, light drinkers, and heavy
drinkers. Am J Cardiol 2006;98:1063–8.
[102] Funatsu K, Yamashita T, Nakamura H. Effect of coffee intake on blood
pressure in male habitual alcohol drinkers. Hypertens Res 2005;28:
521–7.