Supplement Article
Impact of beverage intake on metabolic and
cardiovascular health
Laura Helm and Ian A. Macdonald
This review is based on a presentation that was made at a meeting concerning hydration. It summarizes the epidemiological evidence for selected beverages in relation to cardiovascular and/or metabolic health. The review focuses on tea, cocoa,
milk, orange juice, alcohol, and beverages sweetened with sugars. These beverage
types were chosen because of their widespread consumption, with tea, cocoa, orange juice, and milk being of potential benefit while alcohol and sugars may be
detrimental. There is reasonably consistent evidence of reduced risk of cardiovascular disease (CVD) in association with high consumption of tea, with the tea
flavonoids appearing to be responsible for these benefits. There is also a growing
evidence base for cocoa flavanols to have beneficial cardiovascular effects. The
bulk of the evidence supporting these conclusions is epidemiological and needs to
be confirmed with randomized controlled trials. Milk is associated with reduced risk
of CVD, particularly in relation to blood pressure, with certain milk tripeptides being
implicated in having effects to reduce angiotensin action. Further work is needed to
confirm these potentially beneficial effects. There is some evidence of potentially
beneficial effects of orange juice on aspects of cardiovascular function, but this is
by no means convincing, and further evidence is needed from randomized controlled trials, together with the elucidation of whether any benefits are linked to the
citrus flavanones or simply to the vitamin C content. While there is some evidence
that red wine may convey some health benefits, there is also clear evidence that alcoholic beverages can have undesirable effects on blood pressure and increase the
risk of CVD. It is possible that low to moderate intakes of alcoholic beverages may
be beneficial. There is some evidence that beverages sweetened with sugars may
contribute to increased energy intake and weight gain, and there is also an indication from longitudinal cohort studies that they are associated with an increased
risk of developing type 2 diabetes. The mechanism of this latter association has not
been explained. In conclusion, there is a substantial amount of epidemiological evidence for benefits of tea and cocoa in relation to cardiovascular health. There is a
growing literature describing randomized controlled trials, but more evidence is
needed. Potential cardiovascular and metabolic health benefits of milk and orange
juice needs further investigation. The associations of higher alcohol intakes and
consumption of beverages sweetened with sugars and their increased health risks
are of concern, and more attention should be focused on this area.
Affiliation: L. Helm and I.A. Macdonald are with the School of Life Sciences, University of Nottingham Medical School, Nottingham, United
Kingdom.
Correspondence: I.A. Macdonald, School of Life Sciences, University of Nottingham Medical School, Queen’s Medical Centre, Nottingham,
NG7 2UH, UK. E-mail: [email protected]. Phone: þ44-1158230100.
Key words: cardiovascular risk, cocoa, flavanols, milk tripeptides, orange juice, sugar-sweetened beverages, tea.
C The Author(s) 2015. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For
V
Permissions, please e-mail: [email protected].
120
doi: 10.1093/nutrit/nuv049
Nutrition ReviewsV Vol. 73(S2):120–129
R
INTRODUCTION
There is continuing interest in the potential beneficial
or detrimental effects of various beverages on health. In
particular, claims have been made that tea and cocoa affect cardiovascular function and reduce the risk of cardiovascular disease (CVD), while positive and negative
effects of orange juice and milk have been reported.
While alcohol appears to increase the risk of high blood
pressure, red wine has been linked with reduced risk of
CVD. There is also extensive interest in the potential
risk to health associated with the consumption of beverages sweetened with various types of sugars, commonly
referred to as sugar-sweetened beverages. This is included in the ongoing review of the Effects of Dietary
Carbohydrates on Health being undertaken by a working group of the UK Scientific Advisory Committee on
Nutrition, which is chaired by one of the authors
(I.A.M.). The present review focuses on systematic reviews of cohort/epidemiological studies and of randomized controlled trials (RCTs), with individual studies
only being reported if there is an absence of such reviews, if they have been published since the most recent
review, or if they are linked to possible mechanisms explaining any potential health benefits. This review is not
meant to be comprehensive; it is based on a presentation made at a Workshop on Hydration, which was intended to give an overview of different beverages and
their relationship to cardiometabolic health.
CARDIOVASCULAR HEALTH
Tea
There has been interest in the potential health benefits
of tea for many years, not least because it is one
of the most popular beverages consumed worldwide
(Table 1).
Peters et al.1 reported a meta-analysis of 10 cohort
and 7 case-controlled studies from 1966 to 2001 concerning the relationships between tea intake and stroke,
myocardial infarction, and coronary heart disease.
Overall, tea was associated with a reduction in CVD
risk, despite 3 studies suggesting that the risk of CVD
actually increased with tea consumption. While various
confounders were taken into account, including the
strength of tea in different geographical regions, there
was no explanation as to why the studies from the
United Kingdom and Australia tended to show increased risk of CVD, whereas European studies showed
reduced risk. It remains to be determined whether the
types of tea, addition of milk, socioeconomic factors
linked to tea consumption, or other confounders might
contribute to these differences.
More recently, Arab et al.2 conducted a meta-analysis of the links between green and black tea consumption and stroke. This comprised data from 19 496
participants from 1 cross-sectional, 8 cohort, and 2
case-control studies. It appeared that consuming 3 or
more cups of tea a day reduced risk of stroke by 21%,
with no apparent difference between black or green tea.
Wang et al.3 conducted a meta-analysis of observational
studies investigating links between tea intake and coronary artery disease (CAD). They included 13 studies on
black tea (mainly from the United States and Europe)
and 5 on green tea (from China and Japan). No
significant association was found between black tea
consumption and CAD, but there was significant heterogeneity among the studies. By contrast, green tea
consumption was associated with a reduction in risk
of CAD (1 cup/day reducing the risk by 10%). While
these observations are interesting, the possibility of
Table 1 Summary of the main reviews linking tea consumption with cardiovascular disease events and risk markers
Reference
Peters et al. (2001)1
Arab et al. (2009)2
Wang et al. (2011)3
Taubert et al. (2007)5
Hooper et al. (2008)6
Zheng et al. (2011)7
Ras et al. (2011)8
Hartley et al. (2013)9
Observation
Tea associated with reduced cardiovascular disease
risk
3 or more cups/day reduced stroke risk by 21%
13 studies on black tea showed no association with
coronary artery disease risk
5 studies with green tea; 1 or more cups/day reduced the risk of coronary artery disease by 10%
No link between tea intake and BP in randomized
trials
Acutely, black tea increased SBP by 5.7 mmHg and
DBP by 2.5 mmHg
Green tea reduced LDL cholesterol
Green tea reduced total and LDL cholesterol
Improved brachial artery flow mediated dilation
with black tea
Green tea reduced total cholesterol and BP
Black tea reduced LDL cholesterol and BP
Comment
Studies from the United Kingdom and Australia showed
a trend for increased risk
No obvious difference between green tea and black tea
Possible ethnic differences, as the green tea studies
were from China and Japan
Short duration (4 weeks) may have contributed to the
absence of an effect
Chronic consumption had no effect
No effect on high-density lipoprotein
Of potential importance if the brachial artery is a reliable index of coronary function
Only included studies of >3 months duration
Abbreviations: BP, blood pressure; DBP, diastolic blood pressure; LDL, low-density lipoprotein; SBP, systolic blood pressure.
Nutrition ReviewsV Vol. 73(S2):120–129
R
121
confounding (including geographical/ethnic influences)
limits the wider applicability of the potential benefit.
There is somewhat more limited evidence of the relationship between tea intake and atherosclerosis.
Geleijnse et al.4 reported an analysis of the Rotterdam
Heart Study, which looked at clinically diagnosed atherosclerosis in relation to tea intake. There was a significant negative association between tea intake and severe
aortic atherosclerosis, with an apparent dose response,
such that the relative risk was reduced from 0.54 to 0.31
when tea consumption increased from the range of
125–250 mL/day to >500 mL/day compared to those
who drank less than 125 mL/day.
Many studies of diet and lifestyle effects on CVD
involve risk markers rather than disease endpoints.
Such risk markers include blood lipids, blood pressure,
and endothelial function, and there have been many
studies of the effects of tea intake on these markers.
Taubert et al.5 conducted a meta-analysis of 5 RCTs investigating the association between tea intake and blood
pressure. No significant effect of tea on blood pressure
was found, but the relatively short duration of the trials
(4 weeks) may have contributed to this lack of effect.
Subsequently, Hooper et al.6 systematically reviewed the
evidence from RCTs linking the effect of flavonoid-rich
food sources on CVD risk factors and included black
and green tea studies (8 and 7, respectively). Chronic
consumption of black tea had no effect, while acute
consumption was shown to increase blood pressure—
systolic blood pressure (SBP) by 5.7 mmHg and diastolic BP (DBP) by 2.5 mmHg. This effect seemed to be
independent of the caffeine content of the tea, as some
of the studies controlled for caffeine. By contrast, green
tea significantly reduced low-density lipoprotein (LDL)
cholesterol. The latter observation was supported by a
more recent analysis by Zheng et al.7 who reviewed the
effect of green tea (and green tea extract) on fasting
plasma total cholesterol, LDL cholesterol, and highdensity lipoprotein (HDL) cholesterol in 14 RCTs (1136
participants). Green tea had no effect on HDL cholesterol but it significantly reduced total and LDL
cholesterol.
Ras et al.8 reported a meta-analysis of 9 controlled
human intervention studies investigating the effect of
tea on endothelial function (assessed through the measurement of FMD of the brachial artery). The overall increase in flow mediated dilation (FMD) of the brachial
artery seen with the consumption of 500 mL/day of tea
compared to a placebo was 2.6% of the brachial artery,
which is interesting and potentially beneficial as such
improved endothelial function is an indication of improved coronary vascular function.
The most recent systematic review of the role of
green and black tea in affecting CVD risk markers was
122
undertaken by Hartley et al.9 They limited the RCTs
considered to those longer than 3 months, which resulted in the inclusion of 11 studies (7 looked at the effects of green tea and 4 at black tea). Green tea
consumption significantly reduced total cholesterol and
blood pressure. Black tea reduced LDL cholesterol and
blood pressure.
As summarized in Table 1, there is limited evidence
available to suggest a favorable effect of tea on cardiovascular health. However, caution should be taken due
to the small number of studies included in the review.
To confirm the findings, more long-term studies are required, preferably with RCTs to establish whether there
are effects of tea, particularly whether green or black tea
differ, and whether the addition of milk affects any beneficial effects. It should be noted that since the present
review was completed, several new meta-analyses have
been published showing beneficial effects of tea on
blood pressure.
Cocoa
The potential health benefits of cocoa-based beverages
have been of interest for several centuries, but it is only
relatively recently that RCTs and some epidemiological
studies have examined this scientifically.
Ried et al.10 published a Cochrane Review in 2012
that concluded that flavanol-rich chocolate and cocoa
products have small but statistically significant effects
that lower blood pressure by 2–3 mmHg in the short
term (usually 2 weeks). They also concluded that longerterm studies were needed to draw more meaningful
conclusions about longer-term consumption of flavanolrich cocoa products.
There have been several recent reports of potentially beneficial effects of acute and chronic cocoa flavanol intake from high–flavanol cocoa beverages on blood
pressure and endothelial function. A study by Desideri
et al.11 in older people with mild cognitive impairment,
also published in 2012, showed that 8 weeks of consumption of cocoa flavanol–rich cocoa was accompanied by a reduction in blood pressure and an
improvement in cognitive function. Other work by
West et al.12 showed improvements in endothelial function and arterial stiffness after regular consumption of
cocoa flavanol–rich cocoa (and also after high–cocoa–
flavanol chocolate). These observations build on the
original observations of the Kuna Indians, who are natives of the San Blas Islands in the Caribbean off the
coast of Panama and whose main source of hydration is
a locally produced cocoa beverage. It was found that the
Kuna Indians have low blood pressure and very low
rates of CVD despite high salt intake.13–15 The studies
of the effects of cocoa flavanols on endothelial function
Nutrition ReviewsV Vol. 73(S2):120–129
R
are of particular interest as they point to the potential
mechanisms involving improved endothelial function
due to activation of nitric oxide synthase. Such a mechanism may also underlie the cardiovascular health benefits seen with tea consumption, as flavanols similar to
some found in cocoa are found in tea.
Clearly more work is needed to demonstrate the
mechanisms of any benefits of cocoa for cardiovascular
health. The outcome of studies such as that planned by
National Institutes of Health on the potential effects of
cocoa flavanols on the incidence of cardiovascular
events will be of major interest.
Milk and milk peptides
There is growing interest in the possible cardiovascular
health benefits associated with milk consumption
(Table 2). Some studies have investigated the effects of
milk or milk products, whereas others have isolated tripeptides from milk and investigated their effects.
Clearly, a review related to beverages should focus on
milk, but the possible effects of the tripeptides will also
be considered. One area of concern in relation to milk
is the impact on saturated fat intake. A consideration of
the impact of saturated fat on CVD is beyond the scope
of this review. If there are beneficial effects of the nonlipid components of milk on cardiovascular health, then
this can be exploited by using low-fat or fat-free milk
and milk products.
Blood pressure and risk markers. A number of studies
have investigated the effect of milk, particularly milk tripeptides, on blood pressure and blood lipids. Jauhiainen
and Korpela16 reviewed the epidemiological and experimental studies. The epidemiological studies reported a
link between higher milk intake and lower blood pressure, while the experimental studies showed a decrease
in blood pressure when milk tripeptides were consumed.
This topic was explored further by Xu et al.17 in a
meta-analysis of 9 RCTs studying the effects of
lactotripeptides on blood pressure. The pooling of results from these studies showed a significant reduction
in SBP (4.8 mmHg) and DBP (2.2 mmHg), with greater
effects in hypertensive individuals. Most of the studies
attributed the effects seen to the inhibition of angiotensin converting enzyme. It is notable that all of the studies were from Japan and Finland, and so the wider
applicability of these results to other populations is not
known. Two subsequent studies, Engberink et al.18 in
the Netherlands and Jauhiainen et al.19 in Finland,
failed to show any effect of milk tripeptides on blood
pressure or blood lipids. However, a subsequent metaanalysis by Turpeinen et al.20 of the antihypertensive effects of lactotripeptides in commercial milk products
showed similar overall results to those of Xu et al.17
Nineteen trials were included in the analyses from 1996
to 2010, involving 1500 mildly hypertensive or prehypertensive patients. Overall, SBP was lowered by
4.0 mmHg and DBP by 1.9 mmHg, though not all studies reported a beneficial effect.
A number of other systematic reviews and some
prospective studies that shed light on the potential
health benefits of milk and dairy products have been reported. Ralston et al.21 presented a systematic review of
prospective cohort studies assessing dairy product intake and elevated blood pressure. The review included 5
studies with 45 000 participants and follow-up periods
ranging from 2 to 15 years. There was a significant inverse association between dairy intake and blood pressure, with the highest dairy intake having a relative risk
(RR) of elevated blood pressure of 0.87 (i.e., 13% lower)
when compared to the lowest intake. Of particular interest in this analysis was the observation that the major
benefits appeared to be associated with consumption of
low-fat dairy foods rather than full-fat versions.
Moreover, it appeared that the liquid forms of the dairy
foods (milk, yogurt) were more effective than solid
forms (cheese). Further work is needed to assess
whether the absence of benefits associated with full-fat
dairy products is linked to the effects of fat content on
Table 2 Summary of studies linking milk and milk tripeptide consumption with cardiovascular disease risk markers and
prevalence
Reference
Jauhiainen and
Korpela (2007)16
Observation
Higher milk intake associated with lower BP
Randomized controlled trials show reduced BP with milk
tripeptides
Xu et al. (2008)17
Comment
Most but not all cross-sectional epidemiological studies reported an
inverse association between milk
intake and blood pressure
All studies reviewed were from Japan
and Finland
Beneficial effect not seen in all studies
Milk tripeptides reduced SBP by 4.8 mmHg and DBP by 2.2
mmHg
Turpeinen et al. (2013)20 In mildly hypertensive/prehypertensive patients, milk tripeptides
reduced SBP by 4.0 mmHg and DBP by 1.9 mmHg
Ralston et al. (2012)21
Inverse association between dairy intake and BP; highest intake
Liquid forms of dairy products more
associated with 13% reduced risk of higher BP
effective than solid
Association of milk intake with reduced risk of ischemic heart
Risk 8% lower in highest vs
Elwood et al. (2010)24
disease
lowest milk consumers
Abbreviations: BP, blood pressure; DBP, diastolic blood pressure; SBP, systolic blood pressure.
Nutrition ReviewsV Vol. 73(S2):120–129
R
123
the absorption of minerals such as magnesium and calcium,22 which may have beneficial effects on the cardiovascular system, or to the specific cardiovascular effects
of dietary saturated fats.23
Elwood et al.24 reported an analysis of the association of milk intake with incidence of ischemic heart disease (IHD). Their meta-analysis of 9 cohort studies
showed that the overall risk of IHD in those with the
highest milk consumption compared to those with the
lowest consumption was 0.92. A further analysis of 4
cross-sectional studies (where past milk consumption of
myocardial infarction patients was compared with that
of healthy patients) gave an overall risk of 0.83 of high
milk consumption being associated with IHD, i.e., there
was reduced risk associated with high milk intake. They
also assessed the association of stroke incidence and
milk consumption, performing meta-analyses of 11
studies focusing on thrombo-embolic stroke and 3 on
hemorrhagic stroke. These gave overall risks of 0.79 and
0.75, respectively, for the highest milk consumers vs the
lowest consumers. On the issue of low-fat vs full-fat
milk, it is difficult to draw clear conclusions, as those
who choose to drink low-fat milks are highly likely to
engage in other healthy behaviors; such confounders
will bias results and are difficult to control for.
More recent work. A recent study published after the
census dates for these meta-analyses strengthens the
links between high dairy intake, particularly milk, and
arterial stiffness and blood pressure. Livingstone et al.25
investigated the relationships between dairy foods,
blood pressure, and arterial stiffness (a predictor of cardiovascular events) in 2512 men (aged 45–59 years) at
5-year intervals for 23 years. Compared to those with
the lowest intake of dairy, those with the greatest intake
had reduced arterial stiffness (the augmentation index
was 1.8% lower). More impressively, the SBP in those
with the greatest milk consumption was 10.4 mmHg
lower than in those who never consumed milk. High
milk intakes were associated with lower plasma glucose
levels, and serum triacylglycerol (TAG) levels were
lower with higher intakes of dairy foods.
With regard to possible mechanisms, it has been
suggested that calcium, potassium, and magnesium,
which are known to be required for blood pressure control, are provided in a unique balance in milk. The consumption of these minerals in milk may be a more
effective way of lowering blood pressure than when
they are given as supplements.26,27 Additionally, peptides released from milk during digestion may act to inhibit angiotensin converting enzyme. However, some
doubt has been raised regarding this proposed mechanism of action since these peptides have a very low bioavailability.28 Clearly more work is needed to confirm
124
the beneficial effects of milk and to provide a mechanistic explanation. Studies of the impact of the fat content
of milk and a demonstration that any beneficial effect is
independent of the fat content are essential.
Orange juice
A recent review by Toh et al.29 summarized the evidence
concerning the potential beneficial effects of fruit and
vegetable consumption. They commented that the benefits of citrus flavonoid consumption are inconclusive.
Due to the absence of systematic reviews of the impact of
citrus fruit beverages on cardiovascular health markers,
only individual studies were considered. In an uncontrolled, step-wise increased-dose study in hypercholesterolemic patients, Kurowska et al.30 reported a 21%
increase in HDL cholesterol but only after the third and
highest dose of 750 mL/day. This was also accompanied
by a 30% increase in serum triglycerides. Subsequently,
Cesar et al.31 reported a reduction in LDL cholesterol in
hypercholesterolemic patients (but not in patients with
normal cholesterol) after 8 weeks of consuming 750 mL
orange juice/day. These are clearly large quantities of orange juice with a substantial energy content and not realistic for incorporation into a normal diet.
A recent cross-sectional study looked at the association between orange juice consumption and blood lipids as CVD risk markers.32 The participants were either
of healthy weight or overweight, otherwise healthy (no
diabetes or heart disease), and free from medications.
Forty-one percent of the participants consumed more
than 480 mL of orange juice a day, while the remainder
were nonconsumers. The energy and macronutrient intake and activity levels of the 2 groups did not vary. For
those with a normal lipid profile, the orange juice consumers had lower cholesterol levels than the nonconsumers, while a similar pattern of association, but of
smaller magnitude, was observed in the participants
with increased cholesterol. No significant differences
were seen in body fat or triglyceride levels. A potential
confounder is that the participants in this study were all
employees of an orange juice factory.
A recent study by Morand et al.33 assessed the effects of orange juice and, separately, of its major flavonoid, hesperidin, on cardiovascular risk markers.
Twenty-four healthy overweight men received 500 mL
of orange juice, a control beverage with a placebo, or a
control beverage with hesperidin, each for 4 weeks with
a 3-week washout period in between each intervention
period. There was no effect on overnight fasted
microvascular relaxation or plasma nitric oxide
concentrations. However, orange juice was significantly
more potent in stimulating postprandial microvascular
activity. Orange juice and the hesperidin-control
Nutrition ReviewsV Vol. 73(S2):120–129
R
significantly reduced DBP compared to the placebo, but
no differences were seen in SBP. No significant differences in plasma glucose, insulin, total cholesterol, LDL,
HDL, triglycerides, or inflammatory markers were
found. The mechanisms of these effects remain to be
determined, but the fact that some could be replicated
by hesperidin indicates a possible effect on nitric oxide
synthase by this flavanone.
Research recently undertaken in the laboratory of
one of the current authors (I.A.M.) showed no effect of
4 weeks of orange juice consumption on insulin sensitivity or blood lipids in overweight women. A subsequent study in men showed a reduction in plasma
triacylglycerol concentration after 12 weeks of consumption of 250 mL of orange juice per day, compared
to a small increase in the control group who received a
beverage matched for sugar content but without the citrus flavanones (Simpson et al., unpublished data).
In addition to the potential beneficial effects of citrus flavanones, orange juice also contains vitamin C
(ascorbic acid). There have been a number of studies investigating the association between vitamin C status
(mainly assessed by plasma ascorbic acid levels) and
cardiovascular health. Clearly there are various dietary
sources of vitamin C, and so the following observations
may relate to these other sources as much as to orange
juice. Analysis of the National Health and Nutrition
Examination Survey (NHANES) II data showed
approximately 25% reduced risk of fatal CVD and of
all-cause mortality in patients with normal or elevated
serum ascorbic acid levels compared to those with low
serum levels.34 The European Prospective Investigation
into Cancer and Nutrition (EPIC)-Norfolk study
showed a more dramatic reduction in risk of CVD mortality, with those in the top quintile of plasma ascorbic
acid having a 60% reduced risk compared to those in
the lowest quintile.35 In a prospective case–control
study, risk reduction for coronary heart disease (CHD)
of approximately 33% was observed in those in the
highest quartile of plasma ascorbic acid level compared
to those in the lowest quartile.36
The evidence relating to potential health benefits of
high intakes of vitamin C was reviewed by Frei et al.37
who concluded that the dietary reference value/recommended daily allowance for vitamin C should be increased to 200 mg/day. More work is likely to be needed
before the relevant authorities would reach a similar
conclusion, and it remains to be seen whether this
would, or should, be satisfied by the consumption of citrus juice, whole fruits and vegetables, or supplements.
Overall the evidence of a beneficial effect of orange/
citrus juice on cardiovascular health is rather inconsistent.
Improvements in blood lipids in some studies are not
confirmed in others, while some report improvements in
Nutrition ReviewsV Vol. 73(S2):120–129
R
blood pressure and others do not. More extensive work
with citrus beverages with well-characterized flavanone
contents seems essential to solve these contradictions.
Alcoholic beverages
The relationships between alcoholic beverages, CHD,
and CVD are complex. There is evidence that alcohol
increases blood pressure, but light to moderate consumption of alcohol can increase HDL cholesterol, reduce fibrinogen, and reduce platelet aggregation, which
would, therefore, reduce risk. On the other hand, heavy
intake is associated with an increase in triglycerides and
homocysteine and an increased risk for type 2 diabetes
(T2D). Thus, overall, there appears to be a J-shaped relationship between alcohol intake and disease risk, although it is still contentious whether a little alcohol is
more protective than abstention.
A recent systematic review by Ronksley et al.38 included 84 studies assessing the link between alcohol
consumption and 5 outcomes: CVD mortality, CHD
mortality, CHD morbidity, stroke mortality, and stroke
morbidity. They reported that 2.5–14 g alcohol/day (1
beverage) was associated with 12%–15% reduced risk
compared to not drinking at all for all outcomes. CVD
mortality, stroke incidence, and stroke mortality demonstrated a J-shaped relationship. There was an elevated
risk with intakes over 60 g/day, with an RR of 1.30 for
all-cause mortality. When stroke subtypes were considered, alcohol was associated with reduced risk of ischemic stroke (possibly because of antithrombotic
properties) but increased risk of hemorrhagic stroke.
The authors recognized that as the bulk of the current
evidence is observational, it is difficult to establish causality. An associated review by the same authors (Brien
et al.39) that focused on the effects of alcohol in both observational and intervention studies on plasma lipid
biomarkers associated with CVD showed potentially
beneficial effects of moderate amounts of alcohol (up to
30 g/day) on HDL cholesterol and fibrinogen. No effects
on total cholesterol, LDL cholesterol, or triglycerides
were observed.
The issue of potentially beneficial effects of moderate alcohol intake on CHD risk was addressed by
Hvidtfeldt et al.40 They included 8 prospective studies
(192 067 women and 74 919 men aged 35–89 years) in
the review, none of the participants presented with
CVD at baseline. Food frequency questionnaires and
diet history questionnaires were used to assess alcohol
intakes, and fatal CHD and nonfatal CHD were used as
the outcome measures. An inverse association between
alcohol intake and CHD incidence was seen in all but 1
of the studies. The following confounders were adjusted
for: smoking, body mass index (BMI), education,
125
physical activity, energy intake, poly- and monounsaturated fats, saturated fats, fiber, and cholesterol. A significantly lower risk of CHD in women was seen with
intakes up to 60 g/day and in men with intakes up to
90 g/day. At 30 g/day the RR for women was 0.58 and
for men it was 0.69.
The benefits of a decrease in alcohol intake were
shown by Xin et al.41 They reported a meta-analysis of
RCTs that examined the effect of alcohol reduction on
blood pressure. Fifteen studies were included, with 2234
participants, the majority of whom were male. All of the
studies reported a trend toward a reduction in blood
pressure, with 9 showing a statistically significant reduction in SBP and 8 a reduction in DBP. The overall pooled
effect of alcohol reduction on SBP was 3.31 mmHg and
on DBP it was 2.04 mmHg. There was also a median
reduction in body weight of 0.56 kg.
Thus, the overall conclusion is that reducing alcoholic beverage consumption would reduce cardiovascular risk as represented by blood pressure. However, it is
not yet known whether this requires complete abstention or whether a beneficial reduction in risk would be
achieved by a lower, but not zero, intake.
The potential health benefits of red wine have been
attributed to its flavonoid content. Recent studies have
focused on the effects of supplementation with one of
these compounds, resveratrol, which is of particular interest with regard to metabolic health (see section on
metabolic health below). In relation to cardiovascular
health, Vidavalur et al.42 performed a systematic review
of the role of wine in the reduction of CVD risk. They reported that wine drinkers had higher levels of HDL and
ApoA1 and reduced levels of lipoprotein(a), which are
indices of reduced cardiovascular risk, than nondrinkers.
Moderate red wine consumption was also associated
with reduced platelet aggregation (possibly due to reduced prostanoid synthesis), reduced fibrinogen, and increased fibrinolytic activity. Some of the CVD benefits
may occur because red wine causes the stimulation of nitric oxide-dependent signaling. The researchers also reported evidence of beneficial anti-oxidant and antiinflammatory effects of red wine, but similar claims have
been made for tea and cocoa flavanols that have not been
supported by experimental studies.
Chiva-Blanch et al.43 recently summarized the evidence relating to red wine consumption and CVD.
From their study, it is clear that heavy alcohol consumption is associated with increased mortality and
morbidity. By contrast, moderate alcohol intake, especially in the form of beverages that are high in polyphenols, such as red wine, may confer CVD protective
effects, both in healthy volunteers and those with high
CVD risk. Any such beneficial effects of the polyphenols
in red wine would be difficult to translate into practical
126
advice for the public as there is no information available
to the public regarding the polyphenol content of wine,
and it may be difficult to get people to comply with the
recommended moderate consumption.
METABOLIC HEALTH
The main foci of links between dietary factors and metabolic health have concerned risks of developing obesity
and/or T2D. As far as the beverages are concerned,
most of the evidence concerns links with T2D.
Tea
Jing et al.44 reported a meta-analysis of studies assessing
the link between tea consumption and T2D. Of the 9
studies included, 5 showed no significant association, 2
a possible association, 1 an association for green tea
only, and 1 an association only in persons aged <60
years. A secondary stratified analysis indicated that consuming 4 or more cups of tea per day was associated
with a 20% reduced risk of T2D compared to consumption of fewer than 4 cups. Jing et al.44 acknowledged
that the small number of studies, marked variability in
methods, and rather opportunistic analysis in some
studies that were not primarily designed to assess links
between T2D and tea are of some concern. There was
also limited information about the type of tea used, and
there may have been uncontrolled confounding factors
such as socioeconomic status.
This topic was subsequently studied by
Woudenberg,45 using information from the EPIC
Interact study. A total of 26 039 participants from 8 countries were studied. After adjusting for smoking, physical
activity, dietary intake, and BMI, drinking 4 cups of tea a
day compared with 0 cups reduced the risk of T2D by
16%. The risk was also lower (but not significantly so) in
those who drank 1–4 cups a day, showing a linear inverse
relationship. Although this study accounted for numerous variables, there were some that could not be accounted for, including tea type (green/black/with milk)
and brewing time. It is also possible that the inverse relationship found reflects a healthier lifestyle, which was not
effectively adjusted for in the analysis.
Thus, the epidemiological evidence linking tea and
risk of T2D is intriguing but by no means convincing.
Lack of consistency among the studies and the risk of
residual confounding mean that more evidence is
needed before firm conclusions can be drawn.
Cocoa
It has been reported that the cocoa flavanols may be associated with improved insulin sensitivity, although the
Nutrition ReviewsV Vol. 73(S2):120–129
R
original studies involved chocolate rather than a cocoa
beverage.46 By contrast, Muniyappa et al.47 failed to
show any effect of 2 weeks of cocoa ingestion on insulin
sensitivity in hypertensive patients, and it was recently
also showed that 4 weeks of high cocoa flavanol
consumption does not affect insulin sensitivity in overweight, insulin-resistant women when the measurements are made in the morning, 12 hours after the last
cocoa beverage was consumed (Simpson et al., unpublished data). Thus, further work is needed before firm
conclusions can be drawn regarding cocoa and insulin
sensitivity.
Red wine/resveratrol
Potential metabolic effects of red wine are less clear
than its link with a more favorable profile of cardiovascular risk markers. There is interest in the potentially
beneficial metabolic effects of one of the major red wine
flavonoids, resveratrol, and it is now being studied as a
supplement in an attempt to identify potential beneficial effects. An example of the potentially beneficial
effects of resveratrol is provided by Timmers et al.48
They gave resveratrol daily to obese but otherwise
healthy men for 30 days and observed reductions in
plasma glucose, insulin, calculated Homeostatic Model
Assessment (HOMA) (an index of fasting insulin resistance), and triglycerides, together with a reduction in
liver fat content. Interestingly, muscle fat content increased but this did not have the expected effect to increase insulin resistance, as might be expected in the
obese. It remains to be determined whether these potentially beneficial effects of resveratrol are also seen after consumption of red wine itself. At present, it is
not appropriate to conclude that major benefits for
metabolic health can be derived from red wine
consumption.
Sugar-sweetened beverages
There has been ongoing debate for several decades concerning the potential detrimental effects on health
caused by sucrose and other sugars. The move away
from sucrose to greater use of high-fructose corn syrup
in beverages means that the term sugar-sweetened beverages does not just refer to those sweetened with sucrose, but includes all of the mono- and disaccharide
sweeteners and high-fructose corn syrup. Some reviews
of the links between sugars and health have focused on
all dietary sources, i.e., food and beverages, whereas
others have been concerned with beverage sources only.
The World Health Organization released a report on
sugars and health for review in early 2014. However, at
the time the present review was completed, the
Nutrition ReviewsV Vol. 73(S2):120–129
R
final version of this report had not been published. The
World Health Organization’s report was based on 2 systematic reviews of the literature, including one by Te
Morenga et al.49 that focused on body weight and BMI
and was principally concerned with sugars and body
weight but did present information specifically on beverages. More recently, the UK Scientific Advisory
Committee on Nutrition released a draft report on carbohydrates and health, which included information on
sugars and, specifically, on sugar-sweetened beverages.
Most of the information presented in this section is
based on the latter review.
The systematic review by Te Morenga et al.49
showed that if sucrose is consumed as an isoenergetic
substitution for other dietary carbohydrates, then there
is no effect on body weight. However, in ad libitum situations when energy intake is not limited, there is an
increase in body weight with high sucrose intakes.
These major observations related to total sucrose, not
sugar-sweetened beverages per se, but they did have an
additional analysis in children looking at the association
between consumption of beverages sweetened with sugars and overweight. This showed a relative risk of 1.55
for being overweight in those children with a high intake compared to a low intake, which is clearly a potential public health concern.
Greenwood et al.50 identified 9 studies in a total of 11
publications that were concerned with intake of sugarsweetened beverages and the risk of developing T2D.
However, 5 of these studies could not be included in the
meta-analysis, leaving only 4 cohort studies.51–54 The result of this meta-analysis was a statistically significant association between consumption of sugar-sweetened
beverages and the risk of developing T2D, with an increased relative risk of 1.07/100 mL with sugar-sweetened
beverage consumption.
While most reviews of the literature have not distinguished between total sugars and sugars from sweetened beverages in terms of effects on body weight or
food intake, there have been some reports of individual
studies that present information in relation to this. In
adults, the comparison of beverages sweetened with
sugars vs noncaloric sweeteners has shown higher energy intakes with those containing sugars,55,56 demonstrating incomplete compensation for the sugars in
these sweetened beverages. In addition, RCTs in children have shown similar results for an effect of sugarsweetened beverages on weight (and BMI) gain in
growing children. For example, de Ruyter et al.57 conducted a randomized trial comparing these beverages
with an artificially sweetened beverage on BMI over an
18-month period. Because these were growing children
(age at entry ranged from 4 years 10 months to 11 years
11 months), the BMI is expressed as a z-score from the
127
mean BMI at each child’s age. Normal growth would,
thus, be reflected by no change in the z-score over the
18-month period. The artificially sweetened beverage
group showed stability of BMI, with an increase in zscore of only 0.02 standard deviations, whereas the
group consuming beverages with added sugars had an
increase of 0.19 standard deviations (which was statistically significant). It must be noted that this study is by
far the best designed of the RCTs in children that investigated the effects of sugar-sweetened beverages on
weight, although other studies58 have provided data that
would support the observations by de Ruyter et al.57
Thus, there is some evidence that sugar-sweetened
beverages are associated with increased energy intake in
a free-living situation in which energy intake is not being restricted. This may be a risk factor for the development of obesity, more likely as a simple consequence
of increased energy intake rather than any specific
metabolic effect of the sugars consumed. The epidemiological evidence reporting an association between
sugar-sweetened beverages and risk of T2DM is a
concern, but as with other such studies, confounding
cannot be excluded. Nevertheless, it is unlikely that
major new studies will pursue this topic further, so any
advice offered to the public will have to be based on the
evidence presently available.
CONCLUSION
It is clear that potential benefits to health are provided
by tea, cocoa, and milk, with the major effects or associations being related to aspects of cardiovascular health.
One drawback of the cohort studies is the potential for
confounding effects of other variables, e.g., tea drinking
could be indicative of other healthy diet and lifestyle
choices that are not accurately captured in the other
data obtained in these studies. There is some information regarding the possible mechanisms that might mediate any benefits, with alteration in endothelial nitric
oxide release being linked to tea and cocoa flavanols,
and possible reduction in angiotensin converting enzyme activity being linked to some components of milk,
but much more work is needed to fully characterize
these.
While there may be some health benefits associated
with modest consumption of alcohol, particularly red
wine, it is clear that high intakes of alcoholic beverages
are associated with increased blood pressure and increased risk of CVD and death. Further work is needed
to establish whether compounds such as resveratrol in
red wine do mediate a benefit as far as cardiometabolic
health is concerned and, particularly, whether this offsets the potential health detriments associated with the
alcohol in the wine.
128
There is reasonably convincing evidence of potential health benefits associated with vitamin C intakes
above the current dietary recommendations, but it is
too premature to use this to change the recommended
intake levels. Moreover, it is not clear whether orange
juice (or other fruit juices) as a source of this vitamin C
offers increased benefits compared to other dietary
sources. Further work is needed to determine whether
the flavanone content of orange juice confers cardiovascular benefits similar to those seen in relation to tea and
cocoa flavanols.
There is accumulating evidence that beverages
sweetened with sugars are associated with a number of
potential problems in relation to health. The energy
load is likely to be a major contributor to the weight
gain seen in studies in children and the increased
energy intake seen in adults. The association of sugarsweetened beverages with increased risk of T2D is a
major concern and does not seem to be entirely explained by effects on body weight.
Acknowledgments
Funding. I.A.M. was contracted and funded by the
European Hydration Institute. He received financial
reimbursement for travel and accommodation expenses
from the European Hydration Institute and an honorarium was paid to his University by the European
Hydration Institute.
Declaration of interest. The authors have no relevant interests to declare.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Peters U, Poole C, Arab L. Does tea affect cardiovascular disease? A meta-analysis.
Am J Epidemiol. 2001;154:495–503.
Arab L, Liu W, Elashoff D. Green and black tea consumption and risk of stroke: a
meta-analysis. Stroke. 2009;40:1786–1792.
Wang ZM, Zhou B, Wang YS, et al. Black and green tea consumption and
the risk of coronary artery disease: a meta-analysis. Am J Clin Nutr. 2011;93:506–
515.
Geleijnse JM, Launer LJ, Hofman A, et al. Tea flavonoids may protect against atherosclerosis: the Rotterdam Study. Arch Intern Med. 1999;159:2170–2174.
Taubert D, Roesen R, Schomig E. Effect of cocoa and tea intake on blood pressure:
a meta-analysis. Arch Intern Med. 2007;167:626–634.
Hooper L, Kroon PA, Rimm EB, et al. Flavonoids, flavonoid-rich foods, and cardiovascular risk: a meta-analysis of randomized controlled trials. Am J Clin Nutr.
2008;88:38–50.
Zheng XX, Xu YL, Li SH, et al. Green tea intake lowers fasting serum total and LDL
cholesterol in adults: a meta-analysis of 14 randomized controlled trials. Am J Clin
Nutr. 2011;94:601–610.
Ras RT, Zock PL, Draijer R. Tea consumption enhances endothelial-dependent vasodilation; a meta-analysis. PLoS One. 2011;6:e16974.
Hartley L, Flowers N, Holmes J, et al. Green and black tea for the primary prevention of cardiovascular disease. Cochrane Database Syst Rev. 2013;6:Cd009934.
Ried K, Sullivan TR, Fakler P, et al. Effect of cocoa on blood pressure. Cochrane
Database Syst Rev. 2012;8:CD008893
Desideri G, Kwik-Uribe C, Grassi D, et al. Benefits in cognitive function, blood pressure, and insulin resistance through cocoa flavanol consumption in elderly subjects with mild cognitive impairment: the Cocoa, Cognition, and Aging (CoCoA)
study. Hypertension. 2012;60:794–801.
Nutrition ReviewsV Vol. 73(S2):120–129
R
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
West SG, McIntyre MD, Piotrowski MJ, et al. Effects of dark chocolate and cocoa
consumption on endothelial function and arterial stiffness in overweight adults.
Br J Nutr. 2014;111:653–661.
Hollenberg NK, Martinez G, McCullough M, et al. Aging, acculturation, salt intake,
and hypertension in the Kuna of Panama. Hypertension. 1997;29:171–176.
Hollenberg NK, Fisher ND, McCullough ML. Flavanols, the Kuna, cocoa consumption, and nitric oxide. J Am Soc Hypertens. 2009;3:105–112.
McCullough ML, Chevaux K, Jackson L, et al. Hypertension, the Kuna, and the epidemiology of flavanols. J Cardiovasc Pharmacol. 2006;47 (Suppl 2):S103–S109.
Jauhiainen T, Korpela R. Milk peptides and blood pressure. J Nutr. 2007;137
(3 Suppl 2):825s–829s.
Xu JY, Qin PY, Wang LQ, et al. Effect of milk tripeptides on blood pressure: a metaanalysis of randomized controlled trials. Nutrition. 2008;24:933–940.
Engberink MF, Schouten EG, Kok FJ, et al. Lactotripeptides show no effect on human blood pressure: results from a double-blind randomized controlled trial.
Hypertension. 2008;51:399–405.
Jauhiainen T, Niittynen L, Oresic M, et al. Effects of long-term intake of lactotripeptides on cardiovascular risk factors in hypertensive subjects. Eur J Clin Nutr.
2012;66:843–849.
Turpeinen AM, Järvenpää S, Kautiainen H, et al. Antihypertensive effects of bioactive tripeptides—a random effects meta-analysis. Ann Med. 2013;45:51–56.
Ralston RA, Lee JH, Truby H, et al. A systematic review and meta-analysis of elevated blood pressure and consumption of dairy foods. J Human Hypertens.
2012;26:3–13.
Vaskonen T. Dietary minerals and modification of cardiovascular risk factors.
J Nutr Biochem. 2003;14:492–506.
Appel LJ, Brands MW, Daniels SR, et al. Dietary approaches to prevent and treat
hypertension: a scientific statement from the American Heart Association.
Hypertension. 2006;47:296–308.
Elwood PC, Pickering JE, Givens DI, et al. The consumption of milk and dairy foods
and the incidence of vascular disease and diabetes: an overview of the evidence.
Lipids. 2010;45:925–939.
Livingstone KM, Lovegrove JA, Cockcroft JR, et al. Does dairy food intake predict
arterial stiffness and blood pressure in men? Evidence from the Caerphilly prospective study. Hypertension. 2013;61:42–47.
Griffith LE, Guyatt GH, Cook RJ, et al. The influence of dietary and non-dietary calcium supplementation on blood pressure: an updated meta-analysis of randomized controlled trials. Am J Hypertens. 1999;12:84–92.
Massey LK. Dairy food consumption, blood pressure and stroke. J Nutr.
2001;131:1875–1878.
Foltz M, van der Pijl PC, Duchateau GS. Current in vitro testing of bioactive peptides is not valuable. J Nutr. 2010;140:117–118.
Toh JY, Tan VMH, Lim PCY, et al. Flavonoids from fruit and vegetables: a focus on
cardiovascular risk factors. Curr Atheroscler Rep. 2013;15:368–374.
Kurowska, EM, Spence JD, Jordan J, et al. HDL-cholesterol-raising effect of orange
juice in subjects with hypercholesterolemia. Am J Clin Nutr. 2000;72:1095–1100.
Cesar TB, Aptekmann NP, Araujo MP, et al. Orange juice decreases low-density lipoprotein cholesterol in hypercholesterolemic subjects and improves lipid transfer to high-density lipoprotein in normal and hypercholesterolemic subjects. Nutr
Res. 2010;30:689–694.
Aptekmann NP, Cesar TB. Long-term orange juice consumption is associated with
low LDL-cholesterol and apolipoprotein B in normal and moderately hypercholesterolemic subjects. Lipids Health Dis. 2013;12:119.
Morand C, Dubray C, Milenkovic D, et al. Hesperidin contributes to the vascular
protective effects of orange juice: a randomized crossover study in healthy volunteers. Am J Clin Nutr. 2011;93:73–80.
Simon JA, Hudes ES, Tice JA. Relation of serum ascorbic acid to mortality among
US adults. J Am Coll Nutr. 2001;20:255–263.
Khaw KT, Bingham S, Welch A, et al. Relation between plasma ascorbic acid and
mortality in men and women in EPIC-Norfolk prospective study: a prospective
population study. European Prospective Investigation into Cancer and Nutrition.
Lancet. 2001;357:657–663.
Nutrition ReviewsV Vol. 73(S2):120–129
R
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
Boekholdt SM, Meuwese MC, Day NE, et al. Plasma concentrations of ascorbic acid
and C-reactive protein, and risk of future coronary artery disease, in apparently
healthy men and women: The EPIC-Norfolk prospective population study. Br
J Nutr. 2006;96:516–522.
Frei B, Birlouez-Aragon I, Lykkesfeldt J. Authors perspective: what is the optimum
intake of vitamin C in humans? Crit Rev Food Sci Nutrit. 2012;52:815–820.
Ronskley P, Brien SE, Turner BJ, et al. Association of alcohol consumption with selected CVD outcomes: a systematic review and meta-analysis. BMJ. 2011;22:342.
Brien SE, Ronksley PE, Turner BJ, et al. Effect of alcohol consumption on biological
markers associated with risk of coronary heart disease: systematic review and
meta-analysis of interventional studies. BMJ. 2011;342.
Hvidtfeldt UA, Tolstrup JS, Jakobsen MU, et al. Alcohol intake and risk of coronary
heart disease in younger, middle-aged, and older adults. Circulation. 2010;121:
1589–1597.
Xin X, He J, Frontini MG, et al. Effects of alcohol reduction on blood pressure: a
meta-analysis of RCTs. Hypertension. 2001;38:1112–1117.
Vidavalur R, Otani H, Singal PK, et al. Significance of wine and resveratrol in cardiovascular disease: French paradox revisited. Exp Clin Cardiol. 2006;11:217–225.
Chiva-Blanch G, Arranz S, Lamuela-Raventos RM, et al. Effects of wine, alcohol and
polyphenols on cardiovascular disease risk factors: evidences from human studies.
Alcohol. 2013;48:270–277.
Jing Y, Han G, Hu Y, et al Tea consumption and risk of type 2 diabetes: a metaanalysis of cohort studies. J Gen Intern Med. 2009;24:557–562.
Woudenberg V. Tea consumption and incidence of type 2 diabetes in Europe: the
EPIC-InterAct case-cohort study. PLoS One. 2012;7:e36910.
Grassi D, Desideri G, Necozione S, et al. Blood pressure is reduced and insulin sensitivity increased in glucose-intolerant, hypertensive subjects after 15 days of consuming high-polyphenol dark chocolate. J Nutr. 2008;138:1671–1676.
Muniyappa R, Hall G, Kolodziej TL, et al. Cocoa consumption for 2 wk enhances insulin-mediated vasodilatation without improving blood pressure or insulin resistance in essential hypertension. Am J Clin Nutr. 2008;88:1685–1696.
Timmers S, Konings E, Bilet L, et al. Calorie restriction-like effects of 30 days of resveratrol supplementation on energy metabolism and metabolic profile in obese
humans. Cell Metabol. 2011;14:612–622.
Te Morenga L, Mallard S, Mann J. Dietary sugars and body weight: systematic review and meta-analyses of randomised controlled trials and cohort studies. BMJ.
2013;346:e7492.
Greenwood DC, Threapleton DE, Evans CE, et al. Association between sugarsweetened and artificially sweetened soft drinks and type 2 diabetes: systematic
review and dose-response meta-analysis of prospective studies. Br J Nutr.
2014;112:725–734.
Schulze MB, Manson JE, Ludwig DS, et al. Sugar-sweetened beverages, weight
gain, and incidence of type 2 diabetes in young and middle-aged women. JAMA.
2004;292:927–934.
Palmer JR, Boggs DA, Krishnan S, et al. Sugar-sweetened beverages and incidence
of type 2 diabetes mellitus in African American women. Arch Int Med.
2008;168:1487–1492.
Bhupathiraju SN, Pan A, Malik VS, et al. Caffeinated and caffeine-free beverages
and risk of type 2 diabetes. Am J Clin Nutr. 2013;97:155–166.
The InterAct Consortium. Consumption of sweet beverages and type 2 diabetes
incidence in European adults: results from EPIC-InterAct. Diabetologia.
2013;56:1520–1530.
Raben A, Vasilaras TH, Moller AC, et al. Sucrose compared with artificial sweeteners: different effects on ad libitum food intake and body weight after 10 wk of
supplementation in overweight subjects. Am J Clin Nutrit. 2002;76:721–729.
Reid M, Hammersley R, Hill AJ, et al. Long-term dietary compensation for added
sugar: effects of supplementary sucrose drinks over a 4-week period. Brit J Nutrit.
2007;97:193–203.
de Ruyter JC, Olthof MR, Seidell JC, et al. A trial of sugar-free or sugar-sweetened
beverages and body weight in children. N Engl J Med. 2012;367:1397–1406.
Ebbeling CB, Feldman HA, Chomitz VR, et al. A randomized trial of sugar-sweetened beverages and adolescent body weight. N Engl J Med. 2012;367:1407–1416.
129