cardio1

Caffeine and Cardiovascular Health

Introduction

• Caffeine (1,3,7-trimethylxanthine) is a central nervous system (CNS) stimulant alkaloid found in various plants like coffee and cocoa beans, tea leaves, guarana berries, and the kola nut.
• It can be synthetically manufactured for use as a food additive, in dietary supplements, and in over-the-counter or pharmaceutical preparations.
• Synthetic caffeine is identical to intrinsic or plant-derived caffeine.
• Caffeine is the most frequently ingested pharmacologically active food substance globally (IOM, 2014).
• Moderate caffeine consumption of 400 mg/d is not associated with toxicity, cardiovascular effects, effects on bone status and calcium balance, changes in adult behavior, incidence of cancer, or effects on male fertility (Nawrot et al., 2003).
• These conclusions are recognized by Health Canada, the U.S. Food and Drug Administration (FDA, 2012), the European Food Safety Authority (2015), and the U.S. Dietary Guidelines for Americans (USDHHS & USDA, 2015).
• This report evaluates the scientific literature on caffeine relative to possible cardiovascular effects, specifically, effects on:
  • Total cardiovascular disease (CVD)
  • Coronary heart disease (CHD) and acute myocardial infarction (AMI)
  • Arrhythmia
  • Heart failure
  • Sudden cardiac arrest
  • Stroke
  • Blood pressure
  • Hypertension
  • "Other effects": Heart Rate, Cerebral Blood Flow, Cardiac Output, Plasma Homocysteine, Serum Cholesterol, EKG Parameters, Heart Rate Variability, Endothelial/Platelet Function, Plasma & Urine Catecholamines
• An additional area of investigation evaluates possible caffeine tolerance and its effects on some of these endpoints.

Approach and Methodology

• The study identified relevant, high-quality studies in humans from authoritative secondary sources and an updated literature search using the PubMed bibliographic database.
• The updated literature search included reviews published from May 16, 2014, through May 16, 2016, and studies published in 2015 through September 23, 2016.
• Search terms: ((caffeine) AND (blood pressure OR stroke OR cardiovascular disease OR myocardial infarction OR coronary heart disease OR arrhythmia OR ischemic heart disease OR hypertension OR fibrillation OR homocysteine OR cholesterol)) NOT (mouse OR rat)
• Studies for evaluation were identified based on citation by authoritative bodies, appropriate design, adequate study sample size, and appropriate control of potential confounders.
• Particular emphasis was placed on experimental or interventional studies in which exposures could be well-controlled and responses to those exposures carefully measured or monitored.
• Reliance on high-quality observational studies was necessary to assist in evaluating potential effects of prolonged exposure (some 20 years or more) and chronic health effects that cannot be evaluated in an experimental study.
• Observational studies are confounded in that caffeine consumption information was typically assessed at baseline, with follow-up periods up to several years or decades thereafter.
• Caffeine consumption in observational studies is typically assessed by daily intake of tea or coffee, which makes it difficult to separate the potential effects of caffeine from other ingredients.
• Both coffee and tea contain a variety of other chemicals, particularly polyphenols, which may contribute to the beneficial effects attributed to these beverages.
• Diterpenes in coffee have been implicated in the effects of coffee intake on serum cholesterol (Bak and Grobbee, 1989; Stensvold et al., 1989; Godos et al., 2014).
• Following the identification of potentially relevant studies, these were reviewed in more detail to determine which studies examined the relationship between caffeine dose and relevant cardiovascular effects.
• Data was extracted to assess how the occurrence of these effects varies in incidence/severity with caffeine dose and duration of exposure among subpopulations of interest.
• More than 300 studies were included in the database of pertinent studies.

Background

• Around 85% of the US population consumes at least one caffeinated beverage per day.
• The mean daily caffeine intake from all beverages was 165 ± 1 mg for all ages combined.
• The per capita 90th percentile caffeine intake peaks at ~600 mg/day among 56-year-olds.
• The historical trends in caffeine intakes remain steady despite the introduction of new sources of caffeine to the marketplace.
• Among caffeine consumers, the greatest source of caffeine in the US is coffee, with smaller amounts coming from carbonated soft drinks (e.g., cola), tea, and energy drinks/shots, and very small amounts from cocoa, chocolate, and other minor sources.
• At existing levels of intake, caffeine can produce a number of expected reversible and transient cardiovascular physiological effects, such as transitory increases in blood pressure.
• A number of studies appear to show beneficial effects of consumption of caffeinated beverages on certain cardiovascular endpoints.
• Caffeine's effects, like nutraceuticals, vary among individuals.
• Individuals habituate and become tolerant to many of caffeine's physiological effects.
• Several genetic polymorphisms that either affect caffeine metabolism or receptor-mediated effects have also been identified as possible contributors to the variability of effect.
• A 'bright line' threshold intake level applicable to the general population that would otherwise induce caffeine's effects cannot be determined.
• The ordinary physiological effects of caffeine at current levels of intake are not known to cause any harm to health.
• The observed physiological effects are transient and reversible and have no known long-term health consequences e they are not adverse.
• Establishing so-called “safe” levels of intake based on non-adverse physiological effects of the most sensitive individuals is not meaningful to protecting public health.
• Unsafe levels of intake correspond to excessive levels, e.g., >100 mg/kg bw/day (more than 6000 mg/ person/day; Boyd et al., 1965) e associated with bona fide adverse effects resulting in acute caffeine toxicity.
• Such levels of caffeine may be achieved with abuse of over-the-counter (OTC) tablets and pure caffeine powder but are not realistically achievable with consumption of caffeinated foods or beverages.
• While a single “bright line” between safe and unsafe intakes (as in a traditional “acceptable daily intake” e ADI) cannot be determined, it is clear that among the general population levels up to 600 mg/day e and in some cases as much as 800 mg/day or more e can be and are consumed every day with no ill-effect.

Definition of an Adverse Health Effect

• It is important to differentiate between a physiological effect and an adverse (harmful) effect.
• Caffeine can cause subtle, reversible physiological effects at relatively low doses, such as transient, reversible, minimally elevated blood pressure, that are clearly not adverse.
• Indeed, many individuals appreciate caffeine's beneficial effects, such as increased alertness or improved concentration, which can be demonstrated experimentally at dose levels at or below 100 mg.
• Effects that are transient, reversible, and do not have long-term detrimental health consequences should not be considered adverse.
• This report does not address “Adverse Event Reports” (AERs) associated with consumption of caffeine-containing products.
• It is not possible to determine cause-and-effect relationships between a particular exposure (such as consumption of a caffeinated beverage) and adverse events (FDA).
• This report discusses both what constitute adverse outcomes (e.g., coronary heart disease) and biomarkers of effect (e.g., blood cholesterol or plasma homocysteine), risk factors for cardiovascular disease.

Possible Variability in Response to Caffeine

• There are a variety of factors that may influence an individual's response to caffeine.
• An individual's response may be influenced by pharmacokinetic factors that affect how rapidly caffeine is absorbed, distributed, metabolized, and eliminated (ADME) after being ingested, or by pharmacodynamics factors that influence the interaction between caffeine and its site(s) of action and the effects of that interaction on the body.
• Some of this variability results from genetic polymorphisms e naturally-occurring mutations in genes either involved in caffeine metabolism (i.e., cytochrome P450 1A2 variants) or in receptor-mediated effects (e.g., adenosine receptors) or non-receptor mediated effects (e.g., low catechol-O-methyltransferase (COMT) activity).
• Caffeine is a competitive antagonist of adenosine at the A1 and A2A receptors, and this antagonistic effect is believed to be responsible for many of the physiological effects of caffeine.

ADME Differences

• Differences in ADME and metabolism, particularly differences in activity of cytochrome P450 1A2 (CYP1A2), may lead to differences in plasma caffeine levels and time course from the same dose.
• Steady-state plasma caffeine concentration may vary widely among individuals given the same dose of caffeine.
• Comparison of slow versus fast metabolizers showed that the different polymorphisms were not associated with significant differences in caffeine pharmacokinetics after consumption of 160 mg among the various arms tested, possibly explained by tolerance to caffeine in the corresponding study subjects.
• The vast majority of ingested caffeine is metabolized, largely in the liver, prior to excretion.
• The metabolic pathways are relatively complex.
• At least 16 metabolites at levels of 0.1% or more of administered caffeine dose may be found in the urine of humans.
• In humans, the principal initial step is 3-demethylation of caffeine (1,3,7-trimethylxanthine) to paraxanthine (1,7-dimethylxanthine), where 72e80% of ingested caffeine follows this route, and paraxanthine plasma levels exceed those of caffeine within 8e10 h of ingestion.
• 3-Demethylation in humans appears to be catalyzed specifically by cytochrome P450 1A2 (CYP1A2).
• There is substantial inter-individual variability of CYP1A2 activity that influences the disposition of a substrate such as caffeine, and these variations may be due to factors such as gender, race, genetic polymorphisms, exposure to enzyme inducers, age, exercise, and pregnancy.
• At least 6 different polymorphic forms of CYP1A2 have been reported (CYP1A21A, CYP1A21D, CYP1A21F, CYP1A21L, CYP1A21V and CYP1A21W).
• Four other CYP isoforms (CYP1A1, CYP2E1, CYP3A, and CYP2D6-Met) also have minor roles in the metabolism of caffeine.
• Caffeine half-lives of 2.5e4.5 h or slightly more were measured in humans at dose levels of greater than 2 mg/kg body weight (bw) and up to 4 mg/kg body weight (bw).
• The ADME profile appears to be age-dependent in rats but not so in humans, except in the very young; half-lives of 50e103, 14.4, and 2.6 h have been observed in premature/newborn, 3e5 month and 5e6 month infants, respectively.
• Caffeine clearance reaches or exceeds adult levels by 5e6 months of age.
• Longer half-lives have been observed in breast-fed than in formula-fed infants, and in women in the last trimester of pregnancy compared with controls.
• Caffeine half-lives have also been found to be as high as 50e160 h in humans with severe liver diseases.

Physiological Differences

• The other major source of variability in response to caffeine for certain effects is differences in the receptors in the brain that may lead to caffeine's physiological CNS effects.
• However, while adenosine receptors play a role in cardiac physiology, we did not identify any studies that examined the role of adenosine receptors, or receptor relevant genetic polymorphisms, in the expression of cardiovascular effects of caffeine.

Age Considerations

• Concerns have been raised in some fora that children may be more sensitive than adults to the effects of caffeine, which would presumably necessitate restrictions on children's exposure to caffeine.
• Other than very young infants, whose metabolic abilities may not be completely developed until they are about six months old, there is little evidence that children and adolescents are inherently any more sensitive than adults when their body weight is taken into consideration.

Habituation/Tolerance/Withdrawal

• Habituation and tolerance to some of the acute effect of caffeine develops with repeated and regular intake.
• While caffeine may result in a slight increase in the blood pressure of naïve individuals, habitual caffeine consumers rapidly develop tolerance and no longer respond to caffeine intake with an increase in blood pressure.
• Tolerance also develops to increases in tension, anxiety, and jitteriness associated with caffeine administration.
• Abrupt discontinuation of caffeine consumption results in mild and transient withdrawal symptoms starting after 12e24 h of abstinence and peaking 20e48 h later, characterized by headache, fatigue, drowsiness, irritability, depressed mood, and anxiety.
• Symptoms of caffeine withdrawal vary considerably between different individuals and are usually not harmful, short-term, and self-limiting.
• While caffeine withdrawal has been added to the American Psychiatric Association's “Diagnostic and statistical manual of mental disorders (5th edition),” it does not meet the criteria for “addiction” (e.g. showing compulsive drug seeking behavior and inability to stop).

Results

Overview of Literature

• Hundreds of studies in humans were identified as potentially relevant, including studies where the exposure of interest was to pure caffeine, coffee, tea, or energy drinks.

• These included 310 studies that were reviewed in detail.

• Of these, 158 experimental and 113 observational studies were included in the final analysis because they included information on caffeine dose and examined the relevant effects.

• This included 8 studies in children and adolescents (infants to 19 years of age).

• All of the studies identified and retrieved are summarized in the associated Excel spreadsheet database, where they are categorized by study type, study population (including age, sex, caffeine consumer type, caffeine sensitivity), endpoint evaluated, study duration, and caffeine dosage studied.

• Assumptions made about the caffeine content of coffee, tea, and body weight:

  • 95 mg caffeine/cup of coffee
  • 45 mg caffeine/cup of tea
  • 1 adult = 70 kg

• The wealth of studies available on caffeine and cardiovascular endpoints is atypical for a food ingredient, and permits a thorough evaluation of caffeine's possible cardiovascular effects in spite of identified study limitations (e.g., a single or a few repeated administrations e mostly 4-weeks or less).

• Our understanding of the potential effects of long-term caffeine consumption is informed primarily by the substantial number of observational epidemiological studies that span multiple years e some, 20 years or more.

• Observational studies generally rely on individual dietary recall (typically 1-day recall) at a snapshot in time, so the long-term measure of exposure is often uncertain, thus limiting the conclusions that can be drawn from them and preventing determination of causation in any association that may be identified.

• While a substantial number of observational studies evaluated higher doses (i.e., >600 mg/day), very few experimental studies examined such doses.

• Neither examined very high doses (i.e., >1200 mg/day) precluding determination of clear adverse effect levels.

Total Cardiovascular Disease (CVD)

• Total cardiovascular disease (CVD), also referred to as heart and blood vessel disease, includes numerous heart and blood vessel problems, many of which are related to atherosclerosis, a condition that develops when plaque builds up in the walls of arteries.
• Some of the major endpoints included within total CVD includes acute myocardial infarction (AMI) or heart attack, stroke, heart failure, arrhythmia, and heart valve problems.
• A total of 19 cohort studies (all observational) with a follow-up time of approximately 3e28 years evaluated the potential relationship between caffeine intake and CVD.
• One study showed statistically significant 1.2e2.8-fold increased risks of CVD among adults that drank 1, 2 or 3 or more cups of coffee per day (95, 190, or 385 mg caffeine or more) with a mean follow-up time of 15 years, compared to those who consumed less than 1 cup/day (<95 mg caffeine; Lindsted et al., 1992).
• Four large cohort studies with follow-up periods ranging from 6 to 20 years demonstrated a statistically significant decreased risk (protective effect) of CVD among participants following consumption of green tea or coffee, or the highest (up to 7 or more cups; >665 mg caffeine/day) intake levels studied, compared to a reference group (typically non-consumers or the lowest caffeine consuming group in the study).
• One of these studies showed a dose-related reduction in total CVD mortality among patients with Type II diabetes mellitus, and a statistically significant protective effect among those who reported consuming 7 or more cups of coffee per day (Bidel et al., 2006).
• Five additional studies demonstrated a protective effect against CVD among those within the lower (up to 400 mg/day), but not higher (greater than 400 mg/day) caffeine consumption categories during follow-up periods that ranged from approximately 8 to 28 years within the studies.
• No statistically significant changes (increased or decreased) in CVD risk compared to the relevant comparison group were observed among tea, cola, or coffee drinkers with caffeine intake between 400 and 600 mg/day during follow-up periods that ranged from approximately 8 to 28 years (Ding et al., 2015; Freedman et al., 2012; Greenberg et al., 2007; Mineharu et al., 2011; Sugiyama et al., 2010).
• When analyzed separately, the protective effects of coffee consumption tended to be observed more consistently, and at higher intakes, among women than men.
• The remaining 9 studies demonstrated no statistically significant changes in CVD risk with caffeine or coffee consumption at the highest intake levels studied e ranging from 100 to 400 mg in three studies, 400e600 mg in four studies, and greater than 600 mg in two studies e compared to non-consumers or the lowest caffeine consumption group in the study with approximately 3e27 years of follow-up.
• Based on the available evidence, caffeine consumption at a variety of intake levels, primarily in the form of tea or coffee, is generally associated with either a statistically significant decreased risk of CVD (in almost half of the available studies) or no statistically significant relationship at all, even at intakes above 600 mg of caffeine per day.
• Evidence from many more, well-conducted studies suggest that caffeine consumption is not associated with an increased risk of CVD, and may even be protective against CVD.
• Observational studies of CVD do not include “pure” caffeine exposures, thus the potential harmful or protective effects of other components within caffeine-containing beverages such as coffee or tea cannot be ruled out.
• These findings likewise do not necessarily rule out any potential association with specific cardiovascular diseases within the total CVD category, which will be explored further in this review.

Coronary Heart Disease (CHD) and Acute Myocardial Infarction (AMI)

• Acute myocardial infarction (AMI) also referred to as a “heart attack,” occurs if the flow of oxygen-rich blood to a section of heart muscle suddenly becomes blocked, leading to death of the heart muscle if blood flow is not quickly restored.
• Most heart attacks occur as a result of coronary heart disease (CHD), a condition in which a waxy substance called plaque builds up inside of the coronary arteries (atherosclerosis) which supply oxygen-rich blood to the heart.
• A total of 50 observational studies evaluated the potential relationship between caffeine intake and CHD/AMI relative risk.
• Thirteen of the 50 studies, with follow-up periods ranging from 3 to 35 years, reported a statistically significant increased risk of CHD or AMI at or above the following caffeine intake categories, primarily from coffee:
  • <100 mg/day: 3 studies (1e2 cups/day)
  • 100-<400 mg/day: 6 studies
  • 400e600 mg/day: 2 studies (only among those in the highest intake groups)
  • >600 mg/day: 2 studies (only among those in the highest intake groups)
• One study reported a U-shaped dose-response relationship for risk of CHD events, where statistically significant increased risks were reported at the lowest (204 mg/day) and highest (722 mg/day) caffeine intake categories, but not at moderate intake (446 mg/day) (Happonen et al., 2004).
• Another study reported an increased risk only among normotensive, but not hypertensive patients.
• Another study reported higher risks of CHD/AMI among those who drank boiled coffee compared to filtered.
• Two additional studies reported an increased risk of CHD or AMI only among individuals with a genotype associated with slow caffeine metabolism (CYP1A21F instead of CYP1A21A) when analyzed separately, or low COMT activity genotype (low catecholamine metabolism) compared to non- or low- consumption of coffee.
• In another study, risk of CHD or AMI was increased among men, but not women.
• Most of the available studies (37 of 50) demonstrated no statistically significant increased risk or a significantly decreased risk (protective effect) of CHD or AMI with caffeine consumption compared to non- or low-consumption.
• Of these 37 studies, 21 studies reported no statistically significant change in risk at any daily intake level investigated compared to non- or low-consumption:
  • 100-<400 mg: 5 studies
  • 400e600 mg: 10 studies
  • >600 mg: 6 studies
• An additional 7 of the 37 studies reported a statistically significant protective effect against CHD or AMI among coffee or tea drinkers at the highest intake level investigated in the study compared to non- or low-consumers:
  • 100-<400 mg: 3 studies
  • >600 mg: 4 studies
• Five studies reported a statistically significant increased risk of CHD or AMI among lower caffeine consumption groups and a null effect among higher intake groups (up to 10 cups of coffee/day e 950 mg caffeine/day e in one study, Palmer et al., 1995) compared to non- or very low consumption groups.
• The four remaining studies of the 37 reported a statistically significant protective effect against CHD or AMI among different caffeine intake (e.g., low intake level, tea vs. coffee) and gender groups compared to non- or low-consumption of caffeine.
• Protective effects against CHD or AMI were typically reported among tea drinkers and lower caffeine intake groups (<100 mg/day or 100-<400 mg/day), with null results typically reported among coffee drinkers or higher caffeine intake groups (generally in the 100e600 mg/day range).
• The majority of the evidence indicates that caffeine intake does not increase the risk of CHD or AMI
• Some evidence suggests that some sensitive subgroups, including those with a specific genotype (e.g., CYP1A2 or COMT variants) may be more susceptible to the potential CVD effects of caffeine (or coffee specifically).
• Of the 13 studies that reported statistically significant increased risks of CHD or AMI at least at the highest intake category studied among coffee drinkers, all involved consumption of coffee as the primary exposure.
• Of those studies that suggest that caffeine consumption is not associated with an increased risk of CHD or AMI, 11 studies reported statistically significant decreased risks of CHD or AMI following caffeine consumption at a variety of intake levels ranging from
• Of the nine studies of tea drinkers with caffeine intakes of up to more than 225 mg/day, none reported a statistically significant increased risk of CHD or AMI, with some reporting statistically significant protective effects against CHD or AMI compared to a reference group (typically non-consumers or the lowest caffeine consuming group in the study).
• Overall, the majority of the studies do not suggest that increasing caffeine consumption is associated with an increased risk of CHD or AMI.

Arrhythmia

• Arrhythmia involves any change from the normal sequence of electrical impulses that controls the beating of the heart; these impulses can happen too fast, too slowly, or erratically e causing fluctuations in the normal heartbeat.
• There are a variety of different types of arrhythmias, with some types more severe than others:
  • Atrial fibrillation: upper heart chambers contract irregularly
  • Bradycardia: slow heart rate
  • Conduction disorders: heart does not beat normally
  • Premature contraction: early heart beat
  • Tachycardia: very fast heart rate
  • Ventricular fibrillation: disorganized contraction of the lower chambers of the heart
  • Atrial flutter: rapidly firing signals that cause the muscles in the atria to contract quickly
• The type of arrhythmia investigated differed from study to study, some of which investigated the effects of caffeine consumption on a variety of arrhythmia subtypes.
• Among the four experimental studies, arrhythmias were reported following caffeine ingestion in only the older intervention study (Dobmeyer et al., 1983).
• More recent experimental studies are not suggestive of a relationship between acute caffeine intake at levels ranging from 227 to 500 mg, and induction of arrhythmias including supraventricular tachycardia (SVT), palpitation, and premature beats.
• The study conducted by Lemery et al. involved 80 patients with SVT, and although SVT was induced in all but three patients, there was no statistically significant difference between groups receiving placebo or caffeine on inducibility, or the cycle length of induced tachycardias.
• Zuchinali et al. conducted a study of 51 patients with pre-existing heart failure, and reported no statistically significant differences in the number of premature beats, or episodes of tachycardia between those who consumed 500 mg of caffeine, or ingested a placebo lactose pill.
• Among the three cohort studies that evaluated the effects of daily caffeine intakes greater than 600 mg/day, none reported an increased risk of atrial fibrillation or flutter (Conen et al., 2010; Frost and Vestergaard, 2005; Mostofsky et al., 2016).
• Conen et al. (2010) also reported a statistically significant decreased (protective) risk of atrial fibrillation among women with caffeine consumption ranging from 217 to 326 mg/day.
• Three other cohort studies that evaluated the effects of daily caffeine intakes between 400 and 600 mg reported either a protective effect against any arrhythmia or a null effect between coffee consumption and atrial fibrillation compared to those with no or low consumption (<2 cups/day; <190 mg caffeine/day).
• In addition to the two studies discussed previously that evaluated the effects of daily consumption of 100 to <400 mg of caffeine, Mattioli et al. (2005, 2011) investigated the probability of the resolution of atrial fibrillation through conversion to sinus rhythm (return to normal heart rhythm) within 48 h from the onset of symptoms.
• In a small case-control study of 116 patients that experienced atrial fibrillation, Mattioli et al. (2005) reported an increased risk of atrial fibrillation, and a decreased probability of conversion to sinus rhythm among participants in the highest intake group (>3 cups of coffee/day; >285 mg caffeine/day) compared to never coffee drinkers.
• Overall, the evidence suggests that caffeine is not associated with an increased risk of incident arrhythmia of varying types.
• While reports of a lower probability of atrial fibrillation resolution has been observed in two small observational studies and an effect on conversion to sinus rhythm following an arrhythmia cannot be excluded, the authors of recent experimental and large prospective studies have concluded that moderate to high intakes of caffeine does not increase the risk of incident arrhythmias and are safe even for those with a history of arrhythmia.
• Several large prospective cohort studies involving caffeine intakes up to more than 600 mg/day did not report an increased risk of arrhythmia, including a report of a protective effect in one study among participants that drank six or more cups of coffee/day.

Heart Failure

• Heart failure, sometimes referred to as congestive heart failure, indicates that the heart isn't pumping blood as well as it should, and the weakened heart can't supply the body's cells with enough blood and oxygen.
• A total of five cohort studies evaluated the potential relationship between caffeine intake and heart failure.
• One study of 7495 Swedish men reported a 17% increased risk of heart failure was reported among those who drank 5 or more cups of coffee per day (>475 mg caffeine/day) compared to no intake after adjustment for age, AMI in brothers or sisters, diabetes, chest pain, smoking, alcohol abuse, high blood pressure, and BMI (Wilhelmsen et al., 2001a).
• No increased risks were reported among those who drank 1 to 4 cups per day (95e380 mg caffeine/day).
• Of the four remaining cohort studies (8608 to 59,490 participants) with approximately 9e19 years of follow-up, no statistically significant increased risks were reported among participants who generally drank five or more cups of coffee per day (400e600 mg caffeine) compared to no or the lowest intake category.
• Wang et al. (2011), who conducted a cohort study in 59,490 Finnish adults, reported a protective effect in women who drank up to 6 cups of coffee per day, while non-statistically significant decreased risks were observed among women who reported 10 or more cups per day.
• Overall, the available evidence suggests that caffeine intake is not associated with an increased risk of heart failure.
• Additionally, the largest cohort study available that investigated this relationship reported a protective effect against heart failure among some study participants.

Sudden Cardiac Arrest

• Sudden cardiac arrest or death (SCA/SCD) is a condition in which the heart suddenly and unexpectedly stops beating, which results in a stoppage of blood flow to the brain and other vital organs.
• A total of three observational studies evaluated the potential relationship between caffeine intake and SCA or SCD.
• In a cohort study of 93,676 postmenopausal women followed-up for 11e16 years, no statistically significant relationship was reported between caffeine intake from coffee or tea up to the highest consumption group (mean of 368 mg/day in the fifth quintile compared to 19 mg/day in the first quintile) and SCD (Bertoia et al., 2013).
• de Vreede-Swagemakers et al. (1999) reported a statistically significant increased risk of SCA among 117 SCA patients (all of whom had CHD) who consumed 10 or more cups of coffee per day (approximately 950 mg of caffeine) compared to no coffee intake, i.e., 144 controls with CHD (age 20e75).
• Some statistically significant increased risks were reported in some sub-analyses, including coffee intake of 10 or more cups (>950 mg caffeine) per day, but these results should be interpreted cautiously given the studies’ limitations (de Vreede-Swagemakers et al., 1999; Weinmann et al., 1997).

Stroke

• Stroke occurs when blood flow to an individual's brain is interrupted, causing cell death within the brain due to a lack of proper oxygenation. A stroke may manifest in two primary ways: via blockage of a blood vessel supplying the brain (ischemic stroke), or via bleeding into and around the brain (hemorrhagic stroke).
• Overall, 31 total studies evaluated the relationship between caffeine consumption and stroke incidence and/or mortality.
• Ten of the 31 stroke-related studies evaluated stroke risk by subtype (Hakim et al., 1998; Kokubo et al., 2013, Kuriyama et al., 2006;Larsson et al., 2008, 2011, 2013; Lopez-Garcia et al., 2009; Mostofsky et al., 2010; Sugiyama et al., 2010; Tanabe et al., 2008; Liebeskind et al., 2015; Loftfield et al., 2015).
• There was no statistically significant relationship between coffee and/or tea consumption at any level of consumption investigated and risk of stroke in 19 of 31 the studies.
• Nine out of 31 studies reported statistically significant decreased risks of stroke in some coffee and tea consumers, though these decreases in risk were not always consistent across consumption groups, study sub-populations or stroke subtype within individual studies (Freedman et al., 2012; Kokubo et al., 2013; Kuriyama et al., 2006; Larsson et al. 2011, 2013; Mineharu et al., 2011; Saito et al., 2015; Tanabe et al., 2008; Loftfield et al., 2015).
• Within these nine studies, the levels of consumption at which investigators reported decreased risks varied, ranging from 190 to 475 mg/day of caffeine from coffee and 180e225 mg/day of caffeine from tea.
• Four of these considering stroke subtypes reported decreased relative risks in the combined stroke categories, but these decreases disappeared when further parsing the population by stroke sub-type (Kokubo et al., 2013; Kuriyama et al., 2006; Larsson et al. 2011, 2013).
• One study reported decreased risk of developing cerebral hemorrhage compared to individuals who drank several cups of a green tea per week or less (Tanabe et al., 2008).
• Three studies out of the 31 total stroke-related studies reported an increased relative risk of stroke associated with some level of coffee consumption (Hakim et al., 1998; Marchioli et al., 1996; Mostofsky et al., 2010).
• One study reported a statistically significant association between consumption of at least 5 cups of coffee per day (>475 mg caffeine) and thromboembolic (i.e., ischemic) stroke (RR = 2.3, 95% CI: 1.4, 4.0) (Hakim et al., 1998).
• Another study, reported an increased risk among those reporting at least 5 cups of coffee per day (>475 mg caffeine, OR ¼ 15.3, 95% CI: 2.4, 97.5) (Marchioli et al., 1996).
• A third study reported that the risk of experiencing a stroke within one hour of coffee consumption was higher compared to the risk of experiencing a stroke during periods of coffee non-consumption (RR = 2.0, 95% CI: 0.4, 2.4) (Mostofsky et al., 2010).
• Overall, the weight of evidence (28 out of 31 studies) suggests that there is no statistically significant association between caffeine consumption and the relative risk of stroke.

Blood Pressure

• “Blood pressure” is the force of blood pushing against the walls of the arteries as the heart pumps blood. Blood pressure is expressed as a ratio of two measures: systolic