Documents Submitted to the Committee (Part 2)

Transcription

Institute of Medicine
Food and Nutrition Board and Board on Health Sciences Policy
Documents Submitted to Committee
Adamson, R. 2013. Letter to Margaret Hamburg, Commissioner of Food and Drugs on behalf of
the ABA (American Beverage Association). July 26.
AHPA (American Herbal Products Association). Adopted 2005, revised 2013. AHPA Caffeine
Labeling Policy.
FDA (Food and Drug Administration). 2013. CFSAN Adverse Events Reporting System
(CAERS) Backgrounder for the Institute Of Medicine (IOM) FDA-Requested Workshop
On Potential Health Hazards of Caffeine In Food and Dietary Supplements and Annex 1
Adverse Event Reports. June 30.
Guggenheim, M.J. 2013. Letter to Margaret Hamburg on behalf of Monster Beverage
Corporation. July 29.
Herrera, D. 2013. Press release and letter to FDA from March 2013, urging FDA action on
Monster, other caffeinated energy drinks. Attorney General’s Office, San Francisco, CA.
June 24.
Lucas, M., E.J. O’Reilly, A. Pan, F. Mirzaei, W.C. Willett, O.I. Okereke, and A. Ascherio. 2013.
Coffee, caffeine, and risk of completed suicide: Results from three prospective cohorts of
American adults. Available online at
http://informahealthcare.com/doi/abs/10.3109/15622975.2013.795243.
Nawrot, P., S. Jordan, J. Eastwood, J. Rotstein, A. Hugenholtz, and M. Feeley. 2003. Effects of
caffeine on human health. Food Addit Contam 20(1):1-30.
Pape, S. 2013. Letter to IOM Planning Committee on Potential Health Hazards Associated with
Consumption of Caffeine in Food and Dietary Supplements on behalf of ABA (American
Beverage Association). July 22.
Peck, J. D., A. Leviton, and L. D. Cowan. 2010. A review of the epidemiologic evidence
concerning the reproductive health effects of caffeine consumption: A 2000-2009 update.
Food & Chemical Toxicology 48(10):2549-2576.
Pinney Associates. 2013. Emergency department visits involving energy drinks and limitations
of the Drug Abuse Warning Network (DAWN). Prepared for the ABA (American
Beverage Association). July 25.
Pomeranz, J. L., C. R. Munsell, and J. L. Harris. 2013. Energy drinks: An emerging public health
hazard for youth. J Public Health Policy 34(2):254-271.
Rotstein, J., J. Barber, C. Strowbridge, S. Hayward, R. Huang, and S. Benrejeb Godefroy. 2013.
Energy Drinks: An Assessment of the Potential Health Risks in the Canadian Context,
International Food Risk Analysis Journal, S. Godefroy, P. Brent, and S. La Vieille (Eds.).
Available online at
http://www.intechopen.com/journals/international_food_risk_analysis_journal/energydrinks-an-assessment-of-the-potential-health-risks-in-the-canadian-context.
15-1
City Attorney Dennis Herrera News Release ForImmediateRelease:
March19,2013
Contact:MattDorsey
(415)554‐4662
Herrera, 18 scientists urge FDA action on Monster, other caffeinated energy drinks City Attorney calls letters an ‘important step we can take at the national level,’ while City’s investigation of Monster Energy Drinks’ business practices continues SANFRANCISCO(March19,2013)—CityAttorneyDennisHerreratodayjoined18scientistsand
publichealthprofessionalsinurgingtheU.S.FoodandDrugAdministrationtotakeprompt
regulatoryactiontoprotectchildrenandadolescentsfromthedangersofhighlycaffeinatedenergy
drinks—includingrequiringmanufacturerstopublishcaffeinecontentonproductlabels.
InseparateletterstoFDACommissionerMargaretHamburgthismorning,Herreraandscientific
expertscitefederallawrequiringthatfoodadditiveslikecaffeinebe“GenerallyRecognizedAs
Safe,”orGRAS,fortheirintendedusebasedonaconsensusofscientificopinion.Contradicting
manufacturers’safetyclaims,thescientists’letterfindsthatthecaffeinelevelsinpopularenergy
drinkssuchasMonsterEnergy,Rockstar,andRedBulldonotmeettheGRASstandardrequiredby
federallawbecausetheyposeseriousriskstopublichealth,especiallyforyoungpeopletowhom
theproductsaremarketed.
Youngconsumersofcaffeinatedenergydrinksareparticularlyvulnerabletothedocumented
healthriskscitedintheletters,includingseizures,cardiacarrhythmia,alteredheartrates,elevated
bloodpressure,sleeplessness,anxietyandchildhoodobesity.AccordingtoFDAdatacitedinthe
letter,consumptionofMonsterEnergyhasbeenimplicatedinthereporteddeathsofatleastfive
individuals,with13additionaldeathsbelievedpossiblylinkedto5‐HourEnergy.
“Caffeine‐dosedenergydrinkswillnevergiveyouwings—buttheymaygiveyoudeadlyhealth
problems,especiallyifyou’reayoungpersontargetedsoaggressivelybymarketers,”saidHerrera.
“Theevidencedoesn’tsupportthemanufacturers’safetyclaims.Tothecontrary,nationally
renownedexpertshavefoundthatenergydrinksposeserioushealthrisks,whichareexacerbated
bymanufacturers’marketingtacticstoyouth.I’mgladtojoinwithrespectedscientists,academics
[MORE]
CityAttorneyDennisHerreraNewsRelease—Page2
andpublichealthprofessionalsnationwideinurgingtheFDAtotakestepstoprotectconsumers.
Thisisanimportantstepwecantakeatthenationallevel,evenasmyofficecontinuesto
investigatethebusinesspracticesofMonsterEnergyDrinkshereinCalifornia.”
InNovember,HerrerainvokedCalifornia’stoughUnfairCompetitionLawtodemandevidencefrom
theCorona,Calif.‐basedMonsterBeverageCorporationtosubstantiateitsmarketingclaimsthatthe
dosagesofcaffeinecontainedinitsproductsare“completelysafe”forconsumptionbyadolescents
andadults.TheCityAttorneyiscontinuinghisinvestigationintoMonsterEnergy,whichisthe
largestmanufacturerinagrowingU.S.energydrinkindustryprojectedreach$19.7billioninsales
thisyear.
###
CIN AND COUNN OF SAN FRANCISCO
OFFICE OF THE CIN AnORNEY
DENNIS J. HERRERA
City Attorney
DIRECT DIAL:
E-MAIL:
(415)554-4748
tara.collins@sfgov.org
March 19,2013
The Honorable Margaret Hamburg, M.D.
Commissioner
U.S. FOOD AND DRUG ADMINISTRATION
U.S. Department of Health and Human Services
10903 New Hampshire Avenue
Silver Spring, MD 20993
Re: The Use of Caffeine in Energy Drinks
Dear Dr. Hamburg:
I write with regard to the attached letter from 18 scientists, academics, clinicians, and public
health professionals who have concluded, based on their expertise and review of published
studies, that the caffeine levels in energy drinks pose serious health and safety risks. Because the
letter concludes that there is no consensus among qualified experts supporting the safety of the
high levels of added caffeine in energy drinks, I urge the Food and Drug Administration
("FDA") to take action to protect consumers from these products.
My office has been investigating the safety of highly caffeinated energy drinks, such as Monster
Energy, as well as the marketing practices and advertising claims made by energy drink
companies. l In the wake of recent reports of illness, injury, and even death allegedly linked to
the consumption of energy drinks, I understand that the FDA is also investigating the safety and
health risks associated with the products.
Because energy drinks are conventional beverages,2 the caffeine added to these drinks is a food
additive that must be generally recognized as safe under the conditions of its intended use (i.e.,
Generally Recognized As Safe ("GRAS")). For a food additive to be safe under this standard,
there must be "a reasonable certainty in the minds of competent scientists that the substance is
not harmful under the intended conditions of use." 21 c.F.R. 170.3(i). Under GRAS guidelines,
See my October 31, 2012 letter to Monster Beverages Corporation, available at
http://www .sfcityattorney .org/Modules/ShowDocumenLaspx ?documentID= I 070.
')
- Many energy drink manufacturers mislabel their products as dietary supplements in an apparent attempt to avoid
the requirements of the GRAS standard. We note that Monster Energy recently announced that it is reclassifying
its drinks as conventional beverages.
I
CITY HALL, ROOM
234 . 1 DR. CARLTON B. GOODLETI PLACE' SAN FRANCISCO. CALIFORNIA 941 02-4682
RECEPTION: (415)554-4700· FACSIMILE: (415)554-4715
CIN AND COUNN OF SAN FRANCISCO .
OFFICE OF TH E C IN AnORNEY
The Honorable Margaret Hamburg, M.D.
Re: The Use of Caffeine in Energy Drinks
March 19,2013
Page 2
the burden is on the manufacturer to prove that (1) an additive is safe for its intended use based
on published scientific literature, and (2) there is a consensus of scientific opinion regarding the
safety of the use of the substance. 21 c.F.R. §§ 170.3, 170.30. The FDA has approved caffeine
as GRAS only for use in cola-type beverages in concentrations no greater than 200 parts per
million, which is substantially less than the amount of caffeine added to energy drinks . 21
C.F.R. § 182.1180.
As the accompanying letter demonstrates, the large amount of added caffeine in energy drinks
does not meet the requirements of the GRAS standard because it is neither safe based on
scientific evidence, nor is there expert consensus regarding its safety. To the contrary, as the
letter makes clear, there is a strong consensus of scientific opinion that the caffeine levels in
energy drinks have not been demonstrated to be safe, but rather pose a serious public health risk,
particularly to young people to whom energy drinks are aggressively marketed.
California's food safety and labeling laws adopt the requirements of the Federal Food, Drug and
Cosmetic Act ("FDCA"), including the GRAS standard. My office has statutory authority to file
suit for violations of California law. However, the harms posed by energy drinks are not limited
to California. Therefore, I urge the FDA to exercise its authority under the FDCA to protect
consumers from the dangerously high levels of caffeine found in many energy drinks.
Thank you for your prompt attention to this matter.
Sincerely,
DJH/msd
Enclosure
March 19, 2013
The Honorable Margaret A. Hamburg, M.D.
Commissioner
Food and Drug Administration
10903 New Hampshire Avenue
Silver Spring, MD 20993
Re: The Use of Caffeine In Energy Drinks
Dear Commissioner Hamburg:
Recent reports of health complications, emergency department visits, injuries, and deaths
related to energy drink consumption have spawned widespread concern among scientists,
health professionals, legislators, state and local law enforcement officials, and consumers
regarding the safety of highly caffeinated energy drinks. As researchers, scientists, clinicians,
and public health professionals who have studied and conducted research on energy drinks, we
are writing this letter to summarize the scientific evidence on this issue and encourage action.
Given the evidence summarized below, we conclude that there is neither sufficient evidence of
safety nor a consensus of scientific opinion to conclude that the high levels of added caffeine in
energy drinks are safe under the conditions of their intended use, as required by the FDA’s
Generally Recognized as Safe (GRAS) standards for food additives. To the contrary, the best
available scientific evidence demonstrates a robust correlation between the caffeine levels in
energy drinks and adverse health and safety consequences, particularly among children,
adolescents, and young adults.
DESCRIPTION OF ENERGY DRINKS AND RELATED PRODUCTS
Energy drinks are a relative newcomer to the U.S. marketplace and have surged in popularity in
recent years, particularly among adolescents. Energy drinks are flavored beverages that contain
added amounts of caffeine as well as other additives such as taurine, guarana (a natural source
of caffeine), and ginseng.1-3
The U.S. energy drink industry has grown rapidly since the drinks were first introduced,3,4 and is
projected to reach $19.7 billion in sales by 2013.2 Between 2006 and 2012, Monster Energy®,
the largest U.S. energy drink manufacturer, tripled its sales.5 As a result of aggressive marketing,
energy drinks are particularly popular among adolescents.4,6,7 As noted in a 2010 study
commissioned by the FDA, a “[e]nergy drinks are typically attractive to young people,” and 65%
of energy drink consumers are 13- to 35-year-olds.8 More recent reports show that 30 to 50% of
This report discusses the mean per capita daily caffeine intake from energy drinks as calculated by
estimates from data provided by the Beverage Marketing Corporation. The mean per capita daily intake
tells us nothing about the number of individuals who are ingesting large quantities of these products. The
report relied on data that is now out of date and made assumptions based on caffeine levels in 16 oz
serving sizes, rather than the new 24 oz sizes. Further, the report also acknowledged that “very limited
reliable information is available of the number and age distribution of regular energy drink consumers”
and “there may be underreporting for young person[s]”.8
a
1
adolescents and young adults consume energy drinks.7,9-11 According to Monitoring the Future,
the federally funded national annual survey of students in grades eight through twelve, 35% of
eighth graders and 29% of both tenth and twelfth graders consumed an energy drink during the
past year, and 18% of eighth graders reported using one or more energy drinks every day.12
Energy drinks vary with respect to caffeine content and concentration.1,13 The caffeine content
of many energy drinks is not disclosed on the product label,2 and in these cases, information
about caffeine content must be derived from Internet sources of unknown validity. In general,
the caffeine concentration of energy drinks is much higher than that of sodas, for which the FDA
has recognized 200 parts per million of caffeine (approximately 71 mg per 12 fl oz serving) as
GRAS.14 By contrast, the most popular energy drinks, like Monster Energy®, contain between
160 and 240 milligrams of caffeine per can. Many energy drinks contain as much as 100 mg of
caffeine per 8 fl oz serving2 with some containing as much as 300 mg per 8 fl oz serving.13 In
addition, many energy drink brands are sold in larger, containers that hold multiple servings (16
to 24 fl oz/473 to 710 mL).1 While some energy drink manufacturers properly classify their
drinks as beverages, others label their beverages as dietary supplements. b
Although some brands of coffee contain amounts of caffeine that exceed the FDA’s established
GRAS levels for soda, energy drinks differ from coffee in three important ways. First, the caffeine
in coffee is naturally occurring, while the caffeine in energy drinks is added by the manufacturer
and is thus subject to regulation by the FDA as a food additive. Second, many energy drinks and
related products containing added caffeine exceed the caffeine concentration of even the most
highly caffeinated coffee.13,15 Third, coffee is typically served hot, tastes bitter, and is consumed
slowly by sipping. By contrast, energy drinks are typically carbonated, sweetened drinks that
are served cold and consumed more rapidly. Indeed, energy drinks are often marketed in a
manner that encourages consumers to ingest large quantities quickly (e.g., “pound down,” “chug
it down” c). Unlike coffee, energy drinks are marketed in a manner designed to appeal to youth
and are highly popular with youth. A scientific review funded by the National Institutes of
Health has concluded that the risk for energy drink overdose is increased by the combination of
marketing that specifically targets youth and the developmental risk-taking tendencies of
adolescents.7
bEnergy
“shots” are a subset of energy drinks that come in smaller containers (usually 1.4 to 3 oz) and
have even higher caffeine concentration than regularly-sized energy drinks. Many contain B vitamins,
taurine, flavoring, and sweeteners. Other “energy products” available for purchase include gel packs,
candies, gum, snacks, energy powders, inhalers, and strips, all containing various amounts of added
caffeine.
cLabels
of Monster Energy® products.
2
HEALTH COMPLICATIONS ASSOCIATED WITH THE CONSUMPTION OF ENERGY DRINKS
We are particularly concerned about the health effects of energy drink consumption by children
and adolescents. Younger individuals tend to have greater sensitivity to a given serving of
caffeine than adults because they are more likely to have a lower body mass and are less likely
have already developed a pharmacological tolerance from regular caffeine consumption. The
American Academy of Pediatrics’ Committee on Nutrition and the Council on Sports Medicine
and Fitness recently concluded that “rigorous review and analysis of the literature reveal that
caffeine and other stimulant substances contained in energy drinks have no place in the diet of
children and adolescents.”16
The Institute of Medicine has similarly recommended that any drinks containing caffeine should
not be sold to children at school.17 Pediatric professionals concur and further state that energy
drinks “are not appropriate for children and adolescents and should never be consumed.”16
Other experts have concluded that children and adolescents should not consume more than 100
mg of caffeine per day,7 less than the amount in a single can of most energy drinks.
With respect to adults, the FDA has noted that consumption of 400 mg of caffeine by healthy
adults in the course of a day is not associated with adverse health effects.18 That standard for
“healthy adults” does not take into consideration that individuals have varying sensitivities to
caffeine.19-24 Moreover, consumption of 400 mg “in the course of the day” is an important
qualification because consumers can ingest 400 mg of caffeine from energy drinks very quickly.
Metabolism of caffeine appears to be non-linear at high doses. In one study using caffeineexperienced human subjects, an increase in caffeine dose from 250 to 500 mg was associated
with significant increases in the half-life as well as a decrease in the clearance of caffeine from
the blood, resulting in higher caffeine levels that were sustained much longer compared with the
lower dose.25 An additional consideration is that the negative effects of caffeine at high blood
levels could be compounded by the accumulation of its metabolites (e.g., paraxanthine,
theophylline, theobromine), which are active stimulants themselves.25,26
Our work as public health professionals has included examination of the surveillance methods
used to track adverse health effects associated with energy drink consumption (e.g., emergency
department visits for caffeine-related cardiac events). Despite widespread use of energy drinks,
there are no systematic data collection methods to ascertain the prevalence of possible adverse
health complications related to energy drinks and related products. Therefore, the following
information likely underestimates the actual prevalence of adverse health effects associated
with these beverages.
Fatalities and Injuries: According to information submitted to the FDA through its voluntary
Adverse Event Reporting System, consumption of Monster Energy® was implicated in the deaths
of five individuals, and reports of 13 deaths have cited the possible involvement of 5-Hour
Energy®.27 The FDA has not disclosed the ages of the deceased individuals in these cases.
However, details reported elsewhere indicate that in one case, a 14-year-old girl reportedly died
of a cardiac arrhythmia induced by caffeine after consuming two 24 oz Monster Energy®
3
beverages over two consecutive days.28 Also reported to the FDA were 21 claims of adverse
reactions, some requiring hospitalization, which were reportedly associated with the
consumption of Red Bull®.29 These reports only refer to three of the energy products on the
market, and of course do not include injuries and deaths that were not voluntarily reported to
the FDA. Also, between October 2010 and September 2011, about half of all calls to the National
Poison Data System for energy-drink-related caffeine toxicity concerned children under 6 years
old. This incidence is far greater than for accidental ingestion of other forms of caffeine.30
Emergency Department Visits: The Drug Abuse Warning Network (DAWN) reports U.S.
emergency department (ED) visits using a probability sampling strategy. DAWN conducted a
special analysis of the data related to energy drink consumption, which revealed a ten-fold
increase in ED visits from 2005 to 2009 (1,128 to 13,114).31 DAWN recently issued an update to
that report which showed that the number of energy-drink-related ED visits doubled between
2007 and 2011, from 10,068 to 20,783.32
Cardiovascular Complications: Caffeine produces a number of cardiac effects, which appear in
a more pronounced manner in caffeine-naïve subjects and in those consuming higher doses of
caffeine. The consumption of highly caffeinated energy drinks has been associated with elevated
blood pressure, altered heart rates, and severe cardiac events in children and young adults,
especially those with underlying cardiovascular diseases. A few studies have examined the
effects of caffeine consumption on heart rate and blood pressure in children and adolescents.33,34
Higher doses of caffeine have been associated with caffeine intoxication, resulting in tachycardia,
elevated blood pressure, vomiting, hypokalemia (from beta-adrenergic stimulation), and cardiac
arrhythmias (atrial flutter, atrial fibrillation, atrioventricular nodal reentrant tachycardia, and
ventricular fibrillation).1,3
A study of young adults found that the consumption of a sugar-free energy drink containing 80
mg of caffeine was associated with changes in platelet and endothelial function great enough to
increase the risk for severe cardiac events in susceptible individuals.35 These findings show how
acute effects of caffeine administration on heart rate might result in cardiovascular events
requiring hospitalization, especially in at-risk youth. Caffeine’s effects on blood pressure have
been found to be more pronounced among African American children than White children.36,37
The consumption of energy drinks before or during exercise might be linked to an increased risk
for myocardial ischemia. In healthy individuals who consume caffeine and then exercise
afterwards, significant reductions in myocardial blood flow have been noted by indirect
laboratory measures.38 Several mechanisms have been postulated to explain this effect,
including the ability of caffeine to block adenosine receptors that modulate coronary vasomotor
tone.38 This vasoconstrictive effect might be more pronounced among caffeine-naïve individuals
or those who acutely ingest higher doses of caffeine, such as are present in energy drinks.
4
Seizures: In addition to cardiac events, cases have been reported of new-onset seizures
attributed to energy drink consumption among 15- to 28-year-olds.39-42 In all of these cases,
seizures ceased after the individuals abstained from consuming energy drinks.
Childhood Obesity: Energy drinks have also been shown to contribute to youth obesity due to
their high calorie and sugar content.7,43 One 24-oz can of Monster Energy® contains 81 grams of
sugar, which is equivalent to 6.75 tablespoons.2 The American Academy of Pediatrics’
Committee on Nutrition reports findings that the consumption of excessive carbohydrate
calories from energy drinks increases risk for pediatric overweight and that “energy drinks have
no place in the diet of children and adolescents.”16 In addition, adolescents are at risk for
increased consumption of high-calorie energy beverages due to marketing claims that they
enhance physical and mental performance and increase energy.13
Other Health Issues: Youth with higher caffeine intake commonly report troubling
neurological symptoms, including nervousness, anxiety, jitteriness, and headache.44-46 In one
review, youth consuming 100 to 400 mg of caffeine daily from dietary sources report jitteriness
and nervousness.44 Studies have also linked high caffeine intake to reduced sleep, poor
academic performance, daytime sleepiness (falling asleep at school), aggressive behavior, and
social and attention problems among youth.47-53 With regard to energy drinks in particular,
studies have shown negative behavioral effects among youth including jitteriness, anxiety, and
dizziness, which might undermine students’ ability to stay on task, focus, and perform well.6
Although many energy drink manufacturers assert that additives such as taurine and B-vitamins
improve physical or cognitive performance, current evidence does not support these claims.54
Finally, energy drinks that have higher titratable acidity levels than sports drinks have been
associated with comparatively more tooth enamel loss.55
Health and Safety Effects of Combining Energy Drinks with Alcohol: Energy drinks also
pose unique dangers when combined with alcohol. Although the FDA and CDC have concluded
that the combination of alcohol and energy drinks is unsafe and poses serious health risks,18,56
the latest available national data from Monitoring the Future indicated that 26% of high school
seniors consumed an alcoholic beverage containing caffeine during the past year.12 Because
individuals who consume energy drinks with alcohol underestimate their true level of alcoholrelated impairment (i.e., a “wide-awake drunk”),57-59 the bulk of scientific evidence suggests that
individuals who combine energy drinks with alcohol are more likely to engage in risky behavior
than if they were only consuming alcohol.60-64 Accordingly, consuming energy drinks mixed with
alcohol is associated with serious alcohol-related consequences such as sexual assault and
driving while intoxicated.60 One study found that individuals who mix alcohol and energy drinks
are more likely to report heavy drinking,65 while another study documented a link between
frequent consumption of energy drinks and increased risk for alcohol dependence among
college students.66
5
CONCLUSION
Based on our own research and our review of the published literature cited herein, we conclude
that there is no general consensus among qualified experts that the addition of caffeine in the
amounts used in energy drinks is safe under its conditions of intended use as required by the
GRAS standard, particularly for vulnerable populations such as children and adolescents. On the
contrary, there is evidence in the published scientific literature that the caffeine levels in energy
drinks pose serious potential health risks, including increased risk for serious injury or even
death. We therefore urge the FDA to take prompt action to protect children and adolescents
from the dangers of highly caffeinated energy drinks, including applying the existing GRAS
standard for sodas to energy drinks and other beverages that contain caffeine as an additive. We
also urge the FDA to require that manufacturers include caffeine content on product labels.
Sincerely,
Amelia M. Arria, Ph.D.
Director
Center on Young Adult Health and Development
University of Maryland School of Public Health
8400 Baltimore Avenue, Suite 100
College Park, MD 20740
aarria@umd.edu
Mary Claire O’Brien, M.D.
Associate Professor
Department of Emergency Medicine
Department of Social Science and Health Policy
Wake Forest School of Medicine
One Medical Center Boulevard
Winston-Salem, NC 27157
mobrien@wakehealth.edu
Roland R. Griffiths, Ph.D.
Professor
Departments of Psychiatry and Neuroscience
Johns Hopkins University School of Medicine
5510 Nathan Shock Drive
Baltimore, MD 21224
rgriff@jhmi.edu
6
Patricia B. Crawford, Dr.P.H., R.D.
Adjunct Professor and Director
Atkins Center for Weight and Health
CE Nutrition Specialist
119 Morgan Hall
University of California
Berkeley, CA 94720
pbcraw@berkeley.edu
Additional Signatories
Kavita Babu, M.D., FACEP, FACMT
Fellowship Director
Division of Medical Toxicology
Assistant Professor
Department of Emergency Medicine
UMass Memorial Medical Center
55 Lake Avenue North
Worcester, MA 01655
kavitambabu@gmail.com
Bruce A. Goldberger, Ph.D.
Professor and Director of Toxicology
Departments of Pathology and Psychiatry
University of Florida College of Medicine
4800 S.W. 35th Drive
Gainesville, FL 32608
bruce-goldberger@ufl.edu
William C. Griffin III, Ph.D., R.Ph.
Research Assistant Professor
Center for Drug and Alcohol Programs
Department of Psychiatry and Behavioral Sciences
Medical University of South Carolina
MSC 861
67 Presidential Street
Charleston, SC 29425
griffinw@musc.edu
7
John P. Higgins, M.D., M.B.A. (Hons), M.Phil., FACC, FACP, FAHA, FACSM, FASNC, FSGC
Associate Professor of Medicine, The University of Texas Health Science Center at Houston
Director of Exercise Physiology, Memorial Hermann Ironman Sports Medicine Institute
Chief of Cardiology, Lyndon B. Johnson General Hospital
Principal Investigator HEARTS (Houston Early Age Risk Testing & Screening Study)
Division of Cardiology
6431 Fannin Street, MSB 4.262
Houston, TX 77030
john.p.higgins@uth.tmc.edu
C. Tissa Kappagoda, M.D.
Professor Emeritus
Heart and Vascular Services
Lawrence J. Ellison Ambulatory Care Center
University of California Davis Health System
4860 Y Street, Suite 0200
Sacramento, CA 95817
ctkappagoda@ucdavis.edu
Steven E. Lipshultz, M.D., FAAP, FAHA
George E. Batchelor Professor of Pediatrics and Endowed Chair in Pediatric Cardiology
Professor of Epidemiology and Public Health
Professor of Medicine (Oncology)
Leonard M. Miller School of Medicine, University of Miami
Chief-of-Staff, Holtz Children's Hospital of the University of Miami-Jackson Memorial Medical Center
Director, Batchelor Children's Research Institute
Member, Sylvester Comprehensive Cancer Center, Miami, Florida
Department of Pediatrics (D820)
University of Miami, Leonard M. Miller School of Medicine
P.O. Box 016820
Miami, Florida 33101
slipshultz@med.miami.edu
Kristine Madsen, M.D., M.P.H., FAAP
Fellow, American Academy of Pediatrics
Assistant Professor
School of Public Health, University of California Berkeley
Department of Pediatrics, University of California San Francisco
King Sweesy and Robert Womack Endowed Chair in Medical Research and Public Health
219 University Hall
Berkeley, CA 94720
madsenk@berkeley.edu
8
Cecile A. Marczinkski, Ph.D.
Assistant Professor
Department of Psychological Science
Northern Kentucky University
349 BEP, 1 Nunn Drive
Highland Heights, KY 41099
marczinskc1@nku.edu
Kathleen E. Miller, Ph.D.
Senior Research Scientist
Research Institute on Addictions
University at Buffalo
1021 Main Street
Buffalo, NY 14203
kmiller@ria.buffalo.edu
Jeffrey Olgin, M.D., FACC
Gallo-Chatterjee Distinguished Professor of Medicine
Professor of Medicine & Chief, Division of Cardiology
University of California San Francisco
505 Parnassus Avenue
Room M-1182A, Box 0124
San Francisco, CA 94143
olgin@medicine.ucsf.edu
Kent A. Sepkowitz, M.D.
Physician
Infectious Disease Service
Department of Medicine, Infection Control
Memorial Sloan-Kettering Cancer Center
1275 York Avenue
New York, NY 10065
sepkowik@mskcc.org
Jennifer L. Temple, Ph.D.
Assistant Professor
University at Buffalo
Departments of Exercise and Nutrition Sciences and Community Health and Behavior
3435 Main Street
1 Farber Hall
Buffalo, NY 14214
jltemple@buffalo.edu
9
Dennis L. Thombs, Ph.D., FAAHB
Professor and Chair
Department of Behavioral & Community Health
EAD 709N School of Public Health
3500 Camp Bowie Boulevard
University of North Texas Health Science Center
Fort Worth, TX 76107
dennis.thombs@unthsc.edu
Charles J. Wibbelsman, M.D.
President
California Chapter 1, District IX
American Academy of Pediatrics
Kaiser Permanente
2200 O'Farrell Street, Teen Clinic
San Francisco, California 94115
charles.wibbelsman@kp.org
Acknowledgements:
Special thanks are extended to Brittany A. Bugbee, Kimberly M. Caldeira, Kaitlin A. Hippen, and
Kathryn B. Vincent.
10
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
Wolk BJ, Ganetsky M, Babu KM. Toxicity of energy drinks. Curr Opin Pediatr.
2012;24(2):243-251.
Heckman MA, Sherry K, Gonzalez de Mejia E. Energy drinks: An assessment of their market
size, consumer demographics, ingredient profile, functionality, and regulations in the United
States. Compr Rev Food Sci Food Saf. 2010;9(3):303-317.
Higgins JP, Tuttle TD, Higgins CL. Energy beverages: Content and safety. Mayo Clin Proc.
2010;85(11):1033-1041.
Blankson KL, Thompson AM, Ahrendt DM, Vijayalakshmy P. Energy drinks: What teenagers
(and their doctors) should know. Pediatr Rev. 2013;34(2):55-62.
Edney A. Monster energy drinks cited in death reports, FDA says. Bloomberg News:
Businessweek. 2012; http://www.businessweek.com/news/2012-10-22/monster-energydrinks-cited-in-death-reports-fda-says. Accessed February 12, 2013.
Pennington N, Johnson M, Delaney E, Blankenship MB. Energy drinks: A new health hazard
for adolescents. J Sch Nurs. 2010;26(5):352-359.
Seifert SM, Schaechter JL, Hershorin ER, Lipshultz SE. Health effects of energy drinks on
children, adolescents, and young adults. Pediatrics. 2011;127(3):511-528.
Somogyi LP. Caffeine intake by the US population. Silver Spring, MD: Food and Drug
Administration; 2010.
Malinauskas BM, Aeby VG, Overton RF, Carpenter-Aeby T, Barber-Heidal K. A survey of
energy drink consumption patterns among college students. Nutr J. 2007;6(1):35-41.
Simon M, Mosher J. Alcohol, energy drinks, and youth: A dangerous mix. San Rafael, CA: Marin
Institute; 2007.
Miller KE. Wired: Energy drinks, jock identity, masculine norms, and risk taking. J Am Coll
Health. 2008;56(5):481-490.
Wadley J. Marijuana use continues to rise among U.S. teens, while alcohol use hits historic
lows. Ann Arbor, MI: University of Michigan; 2011.
Reissig CJ, Strain EC, Griffiths RR. Caffeinated energy drinks-A growing problem. Drug
Alcohol Depend. 2009;99(1-3):1-10.
U.S. Code of Federal Regulations, nr 21CFR-182.1180. Food and Drug Administration. 2012.
McCusker RR, Goldberger BA, Cone EJ. Caffeine content of specialty coffees. J Anal Toxicol.
2003;27(7):520-522.
Committee on Nutrition and the Council on Sports Medicine and Fitness. Sports drinks and
energy drinks for children and adolescents: Are they appropriate? Pediatrics.
2011;127(6):1182-1189.
Institute of Medicine. Nutrition standards for foods in schools : Leading the way toward
healthier youth. Washington, DC: National Academies Press; 2007.
Food and Drug Administration. Warning letter to Phusion Projects Inc. College Park, MD:
Food and Drug Administration; 2010.
Adan A, Prat G, Fabbri M, Sànchez-Turet M. Early effects of caffeinated and decaffeinated
coffee on subjective state and gender differences. Prog Neuropsychopharmacol Biol
Psychiatry. 2008;32(7):1698-1703.
Alsene K, Deckert J, Sand P, de Wit H. Association between A2A receptor gene
polymorphisms and caffeine-induced anxiety. Neuropsychopharmacology. 2003;28(9):16941702.
Yang A, Palmer A, de Wit H. Genetics of caffeine consumption and responses to caffeine.
Psychopharmacology. 2010;211(3):245-257.
11
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
Temple JL, Ziegler AM. Gender differences in subjective and physiological responses to
caffeine and the role of steroid hormones. J Caffeine Res. 2011;1(1):41-48.
Cornelis MC, El-Sohemy A, Campos H. Genetic polymorphism of the adenosine A2A receptor
is associated with habitual caffeine consumption. Am J Clin Nutr. 2007;86(1):240-244.
Sepkowitz KA. Energy drinks and caffeine-related adverse effects. JAMA. 2013;309(3):243244.
Kaplan GB, Greenblatt DJ, Ehrenberg BL, Goddard JE, Cotreau MM, Harmatz JS, Shader RI.
Dose-dependent pharmacokinetics and psychomotor effects of caffeine in humans. J Clin
Pharmacol. 1997;37(8):693-703.
Denaro CP, Brown CR, Wilson M, Jacob P, Benowitz NL. Dose-dependency of caffeine
metabolism with repeated dosing. Clin Pharmacol Ther. 1990;48(3):277-285.
Center for Food Safety and Applied Nutrition. Voluntary and mandatory reports on 5-Hour
Energy, Monster Energy, and Rockstar energy drink. Washington, DC: Food and Drug
Administration; 2012.
Kilar S, Dance S. Family sues energy drink maker over girl's death. The Baltimore Sun. 2012;
http://articles.baltimoresun.com/2012-10-19/health/bs-hs-monster-energy-drink-death20121019_1_energy-drink-monster-energy-monster-beverage-corp. Accessed February 12,
2013.
Center for Food Safety and Applied Nutrition. Voluntary reports on Red Bull energy drink.
Washington, DC: Food and Drug Administration; 2012.
Seifert SM, Seifert SA, Schaechter J, Arheart K, Benson BE, Hershorin ER, Bronstein AC,
Lipshultz SE. Energy drink exposures in the American Association of Poison Control Centers
(AAPCC) National Poison Data System (NPDS) database. Paper presented at: Annual
Meeting of the North American Congress of Clinical Toxicology; 2012; Las Vegas, NV.
Drug Abuse Warning Network. Emergency department visits involving energy drinks.
Rockville, MD: Substance Abuse and Mental Health Services Administration, Center for
Behavioral Health Statistics and Quality; 2011.
Drug Abuse Warning Network. Update on emergency department visits involving energy
drinks: A continuing public health concern. Rockville, MD: Substance Abuse and Mental
Health Services Administration, Center for Behavioral Health Statistics and Quality; 2013.
Temple JL, Dewey AM, Briatico LN. Effects of acute caffeine administration on adolescents.
Exp Clin Psychopharmacol. 2010;18(6):510-520.
Turley KR, Gerst JW. Effects of caffeine on physiological responses to exercise in young boys
and girls. Med Sci Sports Exerc. 2006;38(3):520-526.
Worthley MI, Anisha P, Sciscio Pd, Schultz C, Prashanthan S, Willoughby SR. Detrimental
effects of energy drink consumption on platelet and endothelial function. Am J Med.
2010;123(2):184-187.
Savoca MR, MacKey L, Evans CD, Wilson M, Ludwig DA, Harshfield GA. Association of
ambulatory blood pressure and dietary caffeine in adolescents. Am J Hypertens.
2005;18(1):116-120.
Savoca MR, Evans CD, Wilson ME, Harshfield GA, Ludwig DA. The association of caffeinated
beverages with blood pressure in adolescents. Arch Pediatr Adolesc Med. 2004;158(5):473477.
Higgins JP, Babu KM. Caffeine reduces myocardial blood flow during exercise. Am J Med. in
press.
Calabrò RS, Italiano D, Gervasi G, Bramanti P. Single tonic–clonic seizure after energy drink
abuse. Epilepsy Behav. 2012;23(3):384-385.
Iyadurai SJP, Chung SS. New-onset seizures in adults: Possible association with
consumption of popular energy drinks. Epilepsy Behav. 2007;10(3):504-508.
12
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
Babu KM, Zuckerman MD, Cherkes JK, Hack JB. First-onset seizure after use of an energy
drink. Pediatr Emerg Care. 2011;27(6):539-540.
Trabulo D, Marques S, Pedroso E. Caffeinated energy drink intoxication. BMJ Case Rep.
2011;28(8):712-714.
Clauson KA, Shields KM, McQueen CE, Persad N. Safety issues associated with commercially
available energy drinks. J Am Pharm Assoc. 2008;48(3):e55-e63.
Temple JL. Caffeine use in children: What we know, what we have left to learn, and why we
should worry. Neurosci Biobehav Rev. 2009;33(6):793-806.
Bernstein GA, Carroll ME, Crosby RD, Perwien AR, Go FS, Benowitz NL. Caffeine effects on
learning, performance, and anxiety in normal school-age children. J Am Acad Child Adolesc
Psychiatry. 1994;33(3):407-415.
Heatherley SV, Hancock KMF, Rogers PJ. Psychostimulant and other effects of caffeine in 9to 11-year-old children. J Child Psychol Psychiatry. 2006;47(2):135-142.
Calamaro CJ, Mason TBA, Ratcliffe SJ. Adolescents living the 24/7 lifestyle: Effects of caffeine
and technology on sleep duration and daytime functioning. Pediatrics. 2009;123(6):e1005e1010.
James JE, Kristjansson AL, Sigfusdottir ID. Adolescent substance use, sleep, and academic
achievement: Evidence of harm due to caffeine. J Adolesc. 2011;34(4):665-673.
Pettit ML, DeBarr KA. Perceived stress, energy drink consumption, and academic
performance among college students. J Am Coll Health. 2011;59(5):335-341.
Martin CA, Cook C, Woodring JH, Burkhardt G, Guenthner G, Omar HA, Kelly TH. Caffeine
use: Association with nicotine use, aggression, and other psychopathology in psychiatric
and pediatric outpatient adolescents. ScientificWorldJournal. 2008;8:512-516.
Warzak WJ, Evans S, Floress MT, Gross AC, Stoolman S. Caffeine consumption in young
children. J Pediatr. 2011;158(3):508-509.
Anderson BL, Juliano LM. Behavior, sleep, and problematic caffeine consumption in a
college-aged sample. J Caffeine Res. 2012;2(1):38-44.
Drescher AA, Goodwin JL, Silva GE, Quan SF. Caffeine and screen time in adolescence:
Associations with short sleep and obesity. J Clin Sleep Med. 2011;7(4):337-342.
McLellan TM, Lieberman HR. Do energy drinks contain active components other than
caffeine? Nutr Rev. 2012;70(12):730-744.
Jain P, Hall-May E, Golabek K, Zenia Agustin M. A comparison of sports and energy drinksPhysiochemical properties and enamel dissolution. Gen Dent. 2012;60(3):190-199.
Centers for Disease Control and Prevention. Fact sheets: Caffeinated alcoholic beverages.
Atlanta, GA: Centers for Disease Control and Prevention; 2010.
Ferreira SE, de Mello MT, Pompeia S, de Souza-Formigoni ML. Effects of energy drink
ingestion on alcohol intoxication. Alcohol Clin Exp Res. 2006;30(4):598-605.
Marczinski CA, Fillmore MT, Henges AL, Ramsey MA, Young CR. Effects of energy drinks
mixed with alcohol on information processing, motor coordination and subjective reports of
intoxication. Exp Clin Psychopharmacol. 2012;20(2):129-138.
Arria AM, O'Brien MC. The "high" risk of energy drinks. JAMA. 2011;305(6):600-601.
O'Brien MC, McCoy TP, Rhodes SD, Wagoner A, Wolfson M. Caffeinated cocktails: Energy
drink consumption, high-risk drinking, and alcohol-related consequences among college
students. Acad Emerg Med. 2008;15(5):453-460.
Arria AM, Caldeira KM, Kasperski SJ, O’Grady KE, Vincent KB, Griffiths RR, Wish ED.
Increased alcohol consumption, nonmedical prescription drug use, and illicit drug use are
associated with energy drink consumption among college students. J Addict Med.
2010;4(2):74-80.
13
62.
63.
64.
65.
66.
Miller KE. Alcohol mixed with energy drink use and sexual risk-taking: Casual, intoxicated,
and unprotected sex. J Caffeine Res. 2012;2(2):62-69.
Thombs DL, O'Mara RJ, Tsukamoto M, Rossheim ME, Weiler RM, Merves ML, Goldberger BA.
Event-level analyses of energy drink consumption and alcohol intoxication in bar patrons.
Addict Behav. 2010;35(4):325-330.
Howland J, Rohsenow DJ. Risks of energy drinks mixed with alcohol. JAMA.
2013;309(3):245-246.
Berger LK, Fendrich M, Chen H-Y, Arria AM, Cisler RA. Sociodemographic correlates of
energy drink consumption with and without alcohol: Results of a community survey. Addict
Behav. 2011;36(5):516-519.
Arria AM, Caldeira KM, Kasperski SJ, Vincent KB, Griffiths RR, O'Grady KE. Energy drink
consumption and increased risk for alcohol dependence. Alcohol Clin Exp Res.
2011;35(2):365-375.
14
This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Food and Chemical Toxicology 48 (2010) 2549–2576
Contents lists available at ScienceDirect
Food and Chemical Toxicology
journal homepage: www.elsevier.com/locate/foodchemtox
Review
A review of the epidemiologic evidence concerning the reproductive health effects
of caffeine consumption: A 2000–2009 update
Jennifer David Peck a,*, Alan Leviton b, Linda D. Cowan a
a
b
University of Oklahoma Health Sciences Center, College of Public Health, USA
Harvard Medical School, Children’s Hospital, Neuroepidemiology Unit, One Autumn Street, Boston, MA 02215, USA
a r t i c l e
i n f o
Article history:
Received 22 December 2009
Accepted 10 June 2010
Keywords:
Caffeine
Reproduction
Pregnancy outcomes
Spontaneous abortion
Fetal growth
Birth defects
a b s t r a c t
This review of human studies of caffeine and reproductive health published between January 2000 and
December 2009 serves to update the comprehensive review published by Leviton and Cowan (2002).
The adverse reproductive outcomes addressed in this review include: (1) measures of subfecundity;
(2) spontaneous abortion; (3) fetal death; (4) preterm birth; (5) congenital malformations; and (6) fetal
growth restriction. Methodologic challenges and considerations relevant to investigations of each reproductive endpoint are summarized, followed by a brief critical review of each study. The evidence for an
effect of caffeine on reproductive health and fetal development is limited by the inability to rule out plausible alternative explanations for the observed associations, namely confounding by pregnancy symptoms and smoking, and by exposure measurement error. Because of these limitations, the weight of
evidence does not support a positive relationship between caffeine consumption and adverse reproductive or perinatal outcomes.
Ó 2010 Elsevier Ltd. All rights reserved.
Contents
1.
2.
3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1.
Caffeine exposure assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.
Pregnancy signal phenomenon. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.
Residual confounding by smoking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4.
Precision of results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Subfecundity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.
Time to pregnancy studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.
Fecundability vs. Infertility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.
Semen quality and timing of exposure assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4.
Selection bias in studies of semen quality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.
Contribution of male exposures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.
Review of individual studies of subfecundity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.1.
Assisted reproductive technology (ART) outcomes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.2.
Time to pregnancy studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.3.
Ovulatory infertility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.4.
Sperm quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Spontaneous abortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.
Missing early losses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.
Role of abnormal karyotypes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3.
Pregnancy signal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4.
Proper control selection in case-control studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2550
2550
2552
2552
2553
2553
2553
2553
2553
2553
2553
2553
2553
2554
2554
2554
2555
2555
2556
2556
2556
Abbreviations: ART, assisted reproductive technology; BMI, body mass index; CgA, chromogranin A; CYP1A2, cytochrome P450 1A2; CYP1B1, cytochrome P450 1B1;
CYP2E1, cytochrome P450 2E1; GIFT, gamete intra-fallopian transfer; GST, glutathione S-transferase; HR, hazard ratio; IVF, In vitro fertilization; IUGR, intrauterine growth
restriction; LMP, last menstrual period; OR, odds ratio; NAT2, N-acetyltransferase 2; RR, relative risk; SGA, small for gestational age.
* Corresponding author. Address: University of Oklahoma Health Sciences Center, 801 NE 13th St., Oklahoma City, OK 73104, USA. Tel.: +1 405 271 2229x48053; fax: +1
405 271 2068.
E-mail address: jennifer-peck@ouhsc.edu (J.D. Peck).
0278-6915/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.fct.2010.06.019
2550
J.D. Peck et al. / Food and Chemical Toxicology 48 (2010) 2549–2576
3.5.
3.6.
3.7.
4.
5.
6.
7.
8.
The importance of pregnancy duration (gestational age) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Late recognition of fetal demise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Review of individual studies of spontaneous abortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.7.1.
Cohort studies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.7.2.
Nested case-control studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.7.3.
Case-control studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.7.4.
Caffeine metabolism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.7.5.
Recurrent pregnancy loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.7.6.
Recurrent pregnancy loss and caffeine metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fetal death . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.
4.2.
Heterogeneity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3.
Selection of controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.4.
Late recognition of fetal demise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.5.
Review of individual studies of fetal deaths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Gestational age and preterm birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1.
Heterogeneity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2.
Accurate estimation of gestational age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3.
Review of individual studies of preterm birth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Congenital malformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1.
Heterogeneity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.2.
Biased exposure data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.3.
Relevance of time of exposure to disturbed development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.4.
Outcome ascertainment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.5.
Review of individual studies of congenital malformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fetal growth restriction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.1.
Defining growth restriction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.2.
Accurate estimation of gestational age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.3.
Relevant window of susceptibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.4.
7.5.
Review of individual studies of fetal growth restriction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Discussion/conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.1.
Subfecundity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.2.
Spontaneous abortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.3.
Fetal death . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.4.
Preterm birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.5.
Congenital malformations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.6.
Fetal growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.7.
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Conflict of Interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction
This review of the literature on consumption of caffeine-containing products and reproductive health is an update of the comprehensive report previously published by Leviton and Cowan
(2002). As such, this review is restricted to human studies of caffeine and reproductive health published in English between January 2000 and December 2009. The search strategy consisted of a
Pubmed search using the keywords caffeine, coffee or paraxanthine in combination with the following terms: pregnancy, reproduction, fetal development, miscarriage, spontaneous abortion,
pregnancy loss, fetal death, stillbirth, congenital malformations,
birth defects, fetal growth, growth restriction, growth retardation,
small for gestational age, low birthweight, low birth weight, IUGR,
preterm, fertility, fecundity, time to pregnancy, sperm, semen,
twins, twinning, multiple gestation,or multiple pregnancy. The references cited in all original studies and review papers identified
were also examined to ensure completeness.
The reproductive outcomes addressed in this review are organized into six categories: (1) measures of subfecundity; (2) spontaneous abortion; (3) fetal death; (4) preterm birth; (5) congenital
malformations; and (6) fetal growth. For each topic, we begin by
summarizing study design considerations relevant to investiga-
2556
2556
2556
2556
2557
2558
2558
2559
2559
2560
2560
2560
2560
2560
2560
2561
2561
2561
2562
2564
2564
2564
2564
2565
2565
2568
2568
2568
2568
2568
2568
2572
2572
2572
2573
2573
2573
2573
2573
2573
2573
2574
tions of the specific reproductive endpoint. In keeping with the unique format of the previous summary, we then provide a detailed
critical review of each study, identifying strengths as well as methodological limitations that may influence results and restrict inferences that can be drawn from individual studies. Individual studies
are discussed by topic in chronological order. Table 1 lists the
publications reviewed in this report by reproductive outcome evaluated. Two post-1999 publications (Cnattingius et al., 2000; Grosso
et al., 2001) were reviewed previously in Leviton and Cowan
(2002) and, thus, are not summarized in this review.
We begin with a discussion of general methodological concerns
that should be considered when reviewing studies of consumption
of caffeine-containing products and reproductive health.
1.1. Caffeine exposure assessment
Total caffeine consumption is difficult to measure accurately.
Caffeine exposure can occur from various sources including beverages (coffee, tea, soft drinks), chocolate, and some medications.
Furthermore, the caffeine content of individual beverage servings
varies considerably by method of preparation, product brand, and
cup size (Bracken et al., 2002). The most common justification for
assessing caffeine exposure from coffee alone or from coffee and
2551
Table 1
Chronological list of 2000–2009 publications by reproductive outcome evaluated.
Year
First author
2000
2000
2000
2001
2001
2001
2002
2002
2002
2003
2003
2003
2003
2003
2003
2003
2003
2003
2003
2004
2004
2005
2005
2005
2005
2005
2006
2006
2006
2006
2006
2006
2006
2006
2007
2007
2007
2007
2007
2007
2007
2008
2008
2008
2008
2008
2008
2008
2008
2008
2009
2009
2009
2009
2009
2009
Natsume
Torfs
Zusterzeel
Grosso
Signorello
Wen
Clausson
Klebanoff
Klonoff-Cohen
Balat
Bracken
Giannelli
Horak
Orskou
Rasch
Tolstrup
Tough
Vik
Wisborg
Hassan
Khoury
Bech
Parazzini
Santos
Sata
Sobreiro
Bech
Chiaffarino
Cole
George
Grosso
Karypidis
Matijasevich
Tsubouchi
Bech
Bille
Browne
Diego
Infante-Rivard
Maconochie
Schmid
CARE Study Group
Haugen
Mikkelsen
Mongraw-Chaffin
Ramlau-Hansen
Savitz
Slickers
Weng
Xue
Chavarro
Collier
Johansen
Killick
Miller
Schmidt
Subfecundity/
infertility
Spontaneous
abortion
Fetal death
Gestational age/preterm
birth
Congenital
malformations
Fetal growth
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
tea is that coffee is the predominant source of caffeine exposure
and fewer women report high doses from other beverages, foods
or pharmaceuticals. Regardless of the population proportions consuming large quantities from other sources, consumption from all
sources is important for accurate classification of total caffeine intake for individuals, since low to moderate intake from multiple
sources can result in high cumulative caffeine exposure. Furthermore, because coffee also contains many chemicals other than caffeine, it is difficult to disentangle the potential effects of caffeine
from those that may be attributed to other compounds.
Relying on coffee intake alone would likely result in underestimations of total caffeine exposure. The influence of this measurement error on observed associations would depend on whether
use of caffeine from other sources was more, less or equally
X
X
common among women experiencing the outcome under investigation. Exposure measurement error is commonly assumed to be
similar among those with and without each reproductive disorder, resulting in an underestimation of any coffee/caffeine relationship with the reproduction adversity. This underestimation
(bias toward the null) tends to be predictable only when the
exposure is dichotomous and the association is independent of
other errors (Greenland and Gustafson, 2006). Much of the caffeine literature assesses more than two categories of exposure
and thus, misclassification of caffeine intake could produce a bias
either toward or away from the null, depending on the nature of
the errors.
Other critical aspects of caffeine exposure assessment include
the importance of measuring exposures during the relevant
2552
exposure time window and the need to capture changing intake
patterns throughout pregnancy. Caffeine consumption tends to decrease during the early weeks of pregnancy, coinciding with
increasing pregnancy symptoms and aversions (Gadsby et al.,
1993; Cnattingius et al., 2000; Lawson et al., 2004). Retrospective
reports of caffeine intake collected at a single time point as the
average number of daily servings across a large time span such
as the first trimester or entire pregnancy (typically converted to
caffeine in mg/day) will not accurately characterize true exposure
fluctuations. This type of measurement error is especially relevant
when the critical window of exposure for selected outcomes occurs
during the time interval when consumption patterns are changing
in early gestation.
Although a few studies designed for the purpose of investigating caffeine exposure have implemented detailed improvements
in exposure assessment, variations in caffeine exposure by source,
portion size, brewing method, metabolism and fluctuations over
the course of pregnancy continue to result in exposure misclassification. Furthermore, the comparison of findings across studies is
difficult due to the use of different categories of caffeine intake
and different reference groups.
Self-reported caffeine exposure is not only imprecise, it also
fails to account for variability in rates of degradation. The measurement of caffeine metabolites in biologic fluids provides better
information about biologic dose in part because it reflects individual differences in caffeine metabolism (Klebanoff et al., 1998c), but
such methods are not without limitations. As the major metabolite
of caffeine, serum paraxanthine has a short half-life, ranging from 2
to 5 h in early pregnancy to 10 h in late pregnancy (Aldridge et al.,
1981). Thus, serum paraxanthine concentrations reflect recent
exposures within the day immediately preceding specimen collection. Moreover, studies that incorporate caffeine biomarkers are
typically limited to specimens collected at a single time point.
Therefore, biomarker concentrations would accurately represent
previous caffeine intake patterns only when consumption remains
relatively constant over time. The problem with this assumption is
that caffeine consumption is known to decrease throughout the
first months of pregnancy (Lawson et al., 2004). Thus, biomarker
concentrations may also be susceptible to exposure misclassification when a single measurement is intended to represent longterm or usual patterns of exposure during pregnancy.
The rate at which caffeine is cleared from the body may influence
biologic dose and exposure interval. Caffeine metabolites might be
more important than caffeine in producing a biologic effect. Caffeine
clearance rates differ between individuals and are affected by pregnancy and genetics as well as environmental factors such as cigarette
smoking, drugs and diet (Aldridge et al., 1981; Kalow and Tange,
1991; Carrillo and Benitez, 2000; Lampe et al., 2000). Variations in
caffeine metabolism between individuals are mostly attributed
to differences in cytochrome P450 1A2 (CYP1A2) enzyme activity
(Arnaud, 1994). Assessments of CYP1A2 phenotypes and genotypes
have been used to characterize study populations according to high
or low metabolic activity. A comprehensive review of the metabolic
considerations for caffeine exposure assessment during pregnancy
was published by Grosso and Bracken (2005).
1.2. Pregnancy signal phenomenon
Pregnancy symptoms, including aversions to taste and smells,
nausea, and vomiting are more common in healthy pregnancies
that result in live births and occur less frequently among women
whose pregnancies end in spontaneous abortions (Weigel and
Weigel, 1989; Weigel et al., 2006). This relationship is attributed
to a stronger pregnancy signal produced by higher concentrations
of pregnancy hormones in viable pregnancies (Stein and Susser,
1991; Lawson et al., 2002). Caffeine consumption has been shown
to decrease with increasing pregnancy signal symptoms during the
early weeks of pregnancy (Lawson et al., 2004; Cnattingius et al.,
2000). For example, Lawson et al. (2004) reported that mean onset
of nausea, vomiting and appetite loss occurred between 5 and
6 weeks from the last menstrual period, accompanied by a 59% decrease in caffeine intake from coffee between weeks 4 and 6. Thus,
women experiencing viable pregnancies are more likely to reduce
their caffeine intake in response to the pregnancy signal than
women who go on to have a spontaneous abortion. As a result, reduced caffeine consumption may be a consequence of pregnancy
viability rather than increased consumption causing any reproductive adversity. ‘‘Reverse causation” is the term used to describe
such errors in causal inference.
Control for confounding by pregnancy signal symptoms is critical for caffeine studies of spontaneous abortion and fetal death,
and may have importance for studies of other adverse pregnancy
outcomes. This task, however, is complicated by the difficulty of
measuring pregnancy signal symptoms. Studies that include only
dichotomous indicators for nausea and vomiting fail to capture
the severity, frequency and duration of the symptoms. Furthermore, aversions to specific foods or beverages, which may be
equally or more relevant to decreased caffeine consumption, are
rarely assessed. Studies have revealed that women who decrease
their coffee consumption during early pregnancy commonly attributed these changes to an acknowledged aversion to the taste, smell
or thought of coffee (Lawson et al., 2004); whereas, nausea
reported as number of hours per week was not statistically associated with reduced coffee consumption (Lawson et al., 2002). Thus,
separate assessment of coffee aversion has been recommended for
studies of caffeine and pregnancy outcomes (Lawson et al., 2002;
Lawson and LeMasters, 2006). Efforts to disentangle this complex
relationship will require improved, prospective measurement of
all relevant dimensions of the pregnancy signal.
1.3. Residual confounding by smoking
Most investigators recognize the importance of controlling for
confounding by smoking when evaluating the reproductive effects
of caffeine. Smoking and caffeine use are strongly associated, as
heavier smokers tend to consume more caffeine than others
(Schreiber et al., 1988; Zavela et al., 1990). Furthermore, smoking
is considered a risk factor for many adverse reproductive outcomes
such as infertility, spontaneous abortion, fetal growth restriction,
stillbirth, and preterm birth (Cnattingius, 2004). Controlling for
self-reported smoking status, however, may not provide adequate
control for confounding when smoking is measured inaccurately.
The stigma of smoking during pregnancy may lead to inaccurate
reporting of smoking status or under-reporting of the amount
smoked per day (Ford et al., 1997; Klebanoff et al., 1998b; Lindqvist
et al., 2002). Morrison (1984) described the potential for residual
confounding to occur when the more socially acceptable behavior
of caffeine consumption is more accurately reported than the less
acceptable behavior of smoking tobacco. Because the two behaviors are highly correlated, the more accurately reported caffeine
consumption conveys information about tobacco consumption.
Furthermore, many investigators control only for smoking status
(yes/no) without consideration of amount smoked. Thus, incomplete control for the effects of smoking may explain observed associations commonly attributed to caffeine use. Some authors have
attempted to improve upon these measurements by incorporating
cotinine measurements as biomarkers of nicotine exposure. It is
important, however, to acknowledge that these biochemical markers reflect recent tobacco exposures and may or may not accurately
control for actual smoking patterns during the relevant window of
exposure.
1.4. Precision of results
Measures of association (e.g., odds ratios and relative risks) and
confidence intervals are frequently reported to two decimal places
when, in fact, study precision is rarely sufficient for this degree of
detail to be meaningful. Thus all results discussed in this review
are rounded to one decimal place.
2. Subfecundity
Studies assessing the impact of caffeine on fertility potential
have evaluated a variety of outcomes including time to pregnancy,
infertility, semen quality and selected endpoints of assisted reproductive technologies.
2.1. Time to pregnancy studies
Time to pregnancy studies identify determinants of fecundability, defined as the probability of conception within a non-contracepting menstrual cycle (Baird et al., 1986). Each of the three
studies that evaluated the relationship between caffeine/coffee
exposure and time to conception, none of which were specifically
designed to evaluate caffeine exposures, used a retrospective
time to pregnancy design within a population of pregnant women
(Hassan and Killick, 2004; Cole et al., 2006). The most prominent
limitation of this pregnancy-based approach is that participation
is restricted to those who have achieved a pregnancy (Tingen
et al., 2004). Thus, sub-fecund couples are underrepresented, and
sterile couples are excluded entirely. If exposure is associated with
longer wait times or sterility, more highly exposed sub-fecund
women would be excluded and associations would be underestimated (Spira, 1998; Joffe et al., 2005). Given the retrospective nature of the data collection, these studies are also vulnerable to recall
errors for reports of exposure as well as recollected time to conception. Although recall of time to pregnancy has been demonstrated
to be relatively accurate at the group level over several years (Joffe
et al., 1993), women with longer time to pregnancies would have
to recall their past exposures over a longer time period compared
to women with shorter time to pregnancies. Thus, exposure reports
may be influenced by the outcome of delayed conception. The retrospective time to pregnancy study design is also criticized for lack
of ability to account for timing of intercourse, which may interfere
with the ability to accurately identify cycles at risk of conception.
Prospective time to pregnancy studies that collect data on ovulation and timing of intercourse as couples attempt to conceive are
considered the best available, although not without some limitations, such as pregnancy planning bias, low participation rates
and concerns regarding the generalizability of results from volunteer populations (Tingen et al., 2004).
2.2. Fecundability vs. Infertility
Infertility is defined as failure to conceive within 12 months.
Few exposures would be anticipated to lead to all or nothing effects on fertility. Thus, a continuous measure of fecundability is
more informative than a dichotomous indicator of infertility because it allows for detection of diminished fecundity (Savitz
et al., 2002).
2.3. Semen quality and timing of exposure assessment
Semen characteristics, including volume, sperm concentration,
motility, morphology, and markers of DNA damage in sperm, have
been studied as indicators of male fertility potential. Caffeine studies evaluating semen quality have been mostly cross-sectional;
2553
thus, the timing of caffeine assessment does not coincide with
the relevant window for spermatogenesis, which occurs over
74 days (Nussey and Whitehead, 2001). However, coffee consumption appears to be relatively stable in some populations (Barone
and Roberts, 1996). Thus, caffeine consumption assessed at the
time of semen collection might generally reflect exposure throughout spermatogenesis. Patterns of exposure preceding the window
for spermatogenesis would be relevant if delayed or permanent effects on the spermatogenic cycle were suspected via the mechanism of stem cell disruption (Jensen et al., 2006).
2.4. Selection bias in studies of semen quality
Participation rates in studies of semen quality are typically low,
leading to concerns of selection bias when study volunteers differ
from those who refuse (Cohn et al., 2002). If men with fertility
problems are more motivated to participate in semen studies and
more likely to modify their caffeine intake or other behaviors in
response to their concerns, the association between caffeine and
semen quality would be distorted toward the null value due to
overrepresentation of affected individuals with low exposure.
2.5. Contribution of male exposures
Two studies recognized the importance of including caffeine
exposures of male partners when assessing fecundity endpoints
beyond semen characteristics (Klonoff-Cohen et al., 2002; Cole
et al., 2006). These studies evaluated male and female caffeine consumption separately or the combined monthly caffeine intake for
the couple, but the joint effects of caffeine use in both partners
has not been fully explored.
2.6. Review of individual studies of subfecundity
2.6.1. Assisted reproductive technology (ART) outcomes
Klonoff-Cohen, H., Bleha, J., Lam-Kruglick, P., 2002. A prospective
study of the effects of female and male caffeine consumption on the
reproductive endpoints of IVF and gamete intra-fallopian transfer.
Hum. Reprod. 17, 1746–1754. [Discussed also in 3. Spontaneous
Abortion and 5. Gestational Age and Preterm Birth]
This study evaluated the possible contribution of caffeine in
both partners. This prospective cohort study was conducted among
221 couples undergoing in vitro fertilization (IVF) and gamete intra-fallopian transfer (GIFT) in Southern California between 1993
and 1998. Consumption of caffeine in caffeinated or decaffeinated
coffee, tea, soft drinks, cocoa drinks and chocolate was assessed in
relation to sperm characteristics, number of oocytes retrieved and
fertilized, number of embryos transferred, achieving a pregnancy,
live birth deliveries, miscarriage, multiple gestations, and gestational age at birth.
For women, caffeine consumption was not associated with
number of oocytes retrieved, number of oocytes fertilized, number
of embryos transferred or achieving a clinical pregnancy when
undergoing IVF or GIFT (measures of association not presented).
According to the definition provided by the authors, ‘‘‘not having a live birth’ resulted from either not becoming pregnant or
experiencing a miscarriage”. Among females receiving IVF and
GIFT, associations between not achieving a live birth and caffeine
intake of >50 mg/day compared to 0–2 mg/day during the week
before [OR:3.8 (0.9–15.8)] and the week of the procedure [OR:4.0
(0.5–31.1) were imprecise and not statistically significant. Statistically significant associations were identified for caffeine intake
reported for other time periods further removed from the procedure, including usual lifetime intake and intake during the week
of the initial clinical visit, (OR:3.9 (1.3–11.6) for >50 mg/day over
2554
lifetime; OR:3.8 (1.4–10.7) for >50 mg/day during the week of the
initial clinic visit).
Male caffeine consumption was not associated with sperm
count, motility or morphology or the occurrence of pregnancy
(data not shown by authors). On the other hand, caffeine intake
among males was associated with an increased probability of multiple gestations (presumably mostly twins) in couples who
achieved a pregnancy using IVF and GIFT (for increase of 100 mg/
day OR = 2.2 (1.1–4.4) for usual caffeine intake; OR = 3.0 (1.2–7.4)
for week of initial clinical visit; OR = 2.2 (0.9–5.0) for week before
sperm collection). Only 57 of 71 pregnant couples had complete
information on male caffeine intake (usual and week of initial clinical visit) and confounders and only 45 had complete data for the
assessment during the week before sperm collection.
Strengths of this study include being the first to focus on endpoints occurring within couples receiving assisted reproductive
technology (ART) treatment and the assessment of caffeine intake
across time.
The authors acknowledge the major limitation of the study was
the small sample size, which resulted in limited statistical power
and imprecise measures of association. This study population reported an unusually low level of caffeine exposure, presumably
due to fertility concerns. Thus, the study’s capacity to assess the effects of moderate caffeine intake was limited.
2.6.2. Time to pregnancy studies
Hassan, M.A., Killick, S.R., 2004. Negative lifestyle is associated with
a significant reduction in fecundity. Fertil. Steril. 81, 384–392.
This retrospective study enrolled 2112 pregnant women from
prenatal clinics in the United Kingdom. Several lifestyle factors,
including coffee and tea consumption, were evaluated in relation
to subfecundity measured as time to pregnancy >12 months (typically referred to as ‘‘infertility”) and mean time to pregnancy following cessation of birth control. The study strengths include a
high response rate (99%) and large sample size.
No details regarding the measurement of ‘‘coffee and/or tea intake” were disclosed other than being reported retrospectively in
cups per day (combined). Because a cup of coffee contains substantially more caffeine than a cup of tea and information on soda and
energy drinks was lacking, the combined count of servings would
misclassify total caffeine exposure. Women who consumed the
most coffee and/or tea (P7 cups/day) had a marginally longer
average time to pregnancy (10.4 months, 95% CI 8.1–12.8) compared to ‘‘mild” (66 cups/day) users (8.4 months; 95% CI 7.7–9.1)
(p = 0.1) and a greater odds of subfecundity [OR:1.7 (1.1–2.7)].
When mean time to pregnancy was compared graphically
across five categories of coffee/tea consumption (0, 65, 6–10, 11–
15, and 16–20 cups/day), those with the most extreme intake
(16–20 cups/day) appeared to have a longer time to pregnancy
compared to non-consumers (19.7 vs. 9.6 months). The point estimate for the heaviest intake, however, was imprecise (number of
women consuming 16–20 cups/day not reported).
The study limitations, which include potential recall bias and
exposure misclassification, hinder interpretation.
Cole, D.C., Wainman, B., Sanin, L.H., Weber, J.P., Muggah, H., Ibrahim, S., 2006. Environmental contaminant levels and fecundability
among non-smoking couples. Reprod. Toxicol. 22, 13–19.
This study recruited 41 couples receiving prenatal care during
the third trimester of pregnancy with the goal of evaluating effects
of persistent organic pollutants on fecundability. Participants were
restricted to couples who were not smoking at the time of conception and in which the female partner was 6age 35. According to
the authors’ description, similar numbers of couples were selected
who conceived early (first month) and late (>5 months) as well as
‘‘a few” who conceived in between. Thus, subject selection was
based on the time to conception outcome. As a result, the cohort
design approach to data analysis is questionable and comparisons
of the probability of pregnancy in any given month in the exposed
and unexposed groups are likely to be invalid. For valid comparisons, the subjects in the cohort would need to be selected independently of their outcome status. The lack of details defining caffeine
exposure also raised concerns about the general quality of the retrospective exposure assessment and relevance of the exposure
time frame. Given the methodological concerns, this null study offers no useful information regarding caffeine and fecundability.
Killick, S., Trussell, J., Cleland, K., Moreau, C., 2009. Factors associated with subfertility among women attending an antenatal clinic in
Hull. Hum. Fertil. 12, 191–197.
The retrospective study of subfertility conducted among 2317
pregnant women was not designed to address risk associated with
caffeine intake, but rather aimed to identify the most parsimonious
models predicting time to pregnancy greater than 12 months and
time to pregnancy greater than 24 months. Caffeine intake, defined
as coffee and tea intake of 1–6 cups/week and P7 cups/week was
included among the lifestyle variables assessed in this study. However, caffeine was not retained in the final models which were limited to statistically significant predictors of subfertility. No
measures of association for caffeine were reported.
2.6.3. Ovulatory infertility
Chavarro, J.E., Rich-Edwards, J.W., Rosner, B.A., Willett, W.C., 2009.
Caffeinated and alcoholic beverage intake in relation to ovulatory disorder infertility. Epidemiology 20, 374–381.
Women without a history of infertility in the Nurses Health
Study II cohort (n = 18,555) who became pregnant or tried to
become pregnant during the 1991–1999 follow-up period were
selected for this analysis. Caffeine intake was collected from food
frequency questionnaires administered in 1991 and 1995 to assess
food and beverage intake during the previous year. Ovulatory disorder was reported as a reason for infertility in questionnaires
administered every 2 years. Neither total caffeine intake from all
sources nor frequency of coffee, decaffeinated coffee or tea intake
were associated with infertility due to ovulatory disorder. Two or
more soft drinks per day, however, were positively associated with
risk of ovulatory infertility [OR:1.6 (1.2–2.2), independent of caffeine intake and total energy intake. Non-caffeinated, sugared
and diet soft drinks showed similar associations with ovulatory
disorder infertility, suggesting the association may be due to
chance, components of soft drinks other than caffeine or sugar,
or dietary patterns in which soft drinks are preferentially consumed to the exclusion of nutritious alternatives.
The results offer no evidence to indicate a role for caffeine in the
development of ovulatory infertility. The findings, however, are
limited by the potential for exposure misclassification. Although
the exposure data were collected prospectively, measurement error was possible given the exposure assessment was applied to
pregnancies or pregnancy attempts occurring up to 4 years after
the food frequency questionnaire was administered.
2.6.4. Sperm quality
Horak, S., Polanska, J., Widlak, P., 2003. Bulky DNA adducts in human sperm: relationship with fertility, semen quality, smoking, and
environmental factors. Mutat. Res. 537, 53–65.
This cross-sectional study evaluated correlations between coffee consumption and DNA damage in the semen of 179 men,
approximately half of whom had impaired fertility. Men who drank
>250 ml of coffee per day did not have higher levels of DNA adducts than ‘‘non-drinking or sporadically drinking subjects”. Furthermore, no correlations were observed between amount of
coffee consumed and concentrations of DNA adducts or semen
parameters.
This study has a number of limitations, including lack of detail
about exposure assessment and lack of attention to possible confounding by age or other selection factors. Furthermore, reports
of current coffee consumption may not reflect relevant exposures
during the period of spermatogenesis.
The crude results presented in this study provide no suggestive
evidence of a link between coffee intake and DNA damage in
sperm.
Sobreiro, B.P., Lucon, A.M., Pasqualotto, F.F., Hallak, J., Athayde, K.S.,
Arap, S., 2005. Semen analysis in fertile patients undergoing vasectomy: reference values and variations according to age, length of sexual abstinence, seasonality, smoking habits and caffeine intake. Sao
Paulo Med. J. 123, 161–166.
The purpose of this study was to establish reference standards
for semen characteristics in Brazil and to examine effects of exposures such as caffeine consumption on semen measurements.
Although described as a prospective study, these cross-sectional
data were collected from a population of 500 healthy, fertile men
age 24–63 scheduled for vasectomies. Semen samples were collected in the hospital before the vasectomy. Coffee use was categorized as 0, 1–3, 4–6 and >6 cups per day, but no details regarding
the data collection procedures were provided. Mean values for semen volume, sperm concentration and sperm morphology did not
vary meaningfully across the four categories of coffee use. Mean
sperm motility, however, was observed to increase slightly with
increasing coffee consumption, from 57.1% (sd 16.2) for non-users
to 62.4% (sd 16.0) for men consuming >6 cups/day (p = 0.037).
Thus, no adverse effects of coffee drinking on sperm quality were
observed.
Despite the large study size, the lack of study design details and
the absence of control for confounding limit this study’s contribution to resolving the question of what effects coffee and/or caffeine
may have on reproduction. By restricting their sample to men
seeking a vasectomy (i.e., men who have reason to consider themselves fertile), the authors have eliminated men who might have
limited fertility. Thereby, the authors have reduced their ability
to identify a coffee effect.
Schmid, T.E., Eskenazi, B., Baumgartner, A., Marchetti, F., Young, S.,
Weldon, R., Anderson, D., Wyrobek, A.J., 2007. The effects of male age
on sperm DNA damage in healthy non-smokers. Hum. Reprod. 22,
180–187.
A total of 80 non-smoking men participated in this cross-sectional study evaluating factors associated with DNA damage in
sperm. DNA damage was measured using single-cell electrophoresis (sperm Comet) and reported as the average percentage of DNA
staining outside the area of the sperm nucleus, referred to as % tail
DNA.
Alkaline % tail DNA (indicating single-strand DNA breaks) was
not related to caffeine consumption. Under neutral conditions representing double-stranded DNA breaks, men with the highest caffeine consumption (>308 mg/day) had 19% higher mean % tail DNA
compared to non-users (39.1 (sd 10.5) vs. 32.8 (sd 6.7), p = 0.005).
Neutral % tail DNA did not correlate with semen quality as measured by sperm concentration, total sperm count, motility, and
progressive motility.
The results suggest that DNA damage in sperm may be slightly
increased in healthy men consuming large quantities of caffeine,
but bias due to confounding and exposure misclassification cannot
be ruled out. Although the paper’s focus was on the effects of male
age, age and factors other than total kilocalorie intake and the history of urinary tract infections were not controlled in sub-analyses
of caffeine use.
Ramlau-Hansen, C.H., Thulstrup, A.M., Bonde, J.P., Olsen, J., Bech,
B.H., 2008. Semen quality according to prenatal coffee and present caffeine exposure: two decades of follow-up of a pregnancy cohort. Hum.
Reprod. 23, 2799–2805.
2555
Young-adult sons (n = 343) of women who were members of a
Danish pregnancy cohort were assessed for semen quality and serum hormone concentrations in relation to intrauterine coffee
exposure and current caffeine consumption patterns. Average
maternal coffee exposure during pregnancy was obtained from
questionnaires completed during the 36th week of gestation,
which collected categorical responses as 0–3, 4–7 and P8 cups/
day. At the time of semen and serum collection, the male participants reported their current daily coffee and cola intake. Categorical responses for coffee and cola were each assigned an
estimated caffeine concentration (e.g., 1–4 cups coffee/
day = 250 mg; half to one liter of cola per day = 75 mg), which were
combined to create categories of low (0–25 mg), medium (50–
125 mg) and high (175–1075 mg) caffeine exposure. After adjustment for confounders, including mother’s and son’s smoking, a
non-significant trend toward decreasing mean testosterone (19.2,
17.7 and 17.8 nmol/l for 0–3, 4–7, and P8 cups/day, p = 0.06)
and inhibin B levels (181, 173, and 162 ng/ml; p = 0.09) was observed with increasing prenatal coffee exposure. Current caffeine
consumption was associated with increased adjusted mean testosterone levels (17.9, 19.0, and 20.4 nmol/l for low, medium, and
high caffeine; p = 0.007). No effects on sperm concentration, total
sperm count, semen volume, morphology or motility were observed in adjusted analyses of either prenatal coffee or current caffeine exposures.
Among mothers with high coffee consumption during pregnancy, a higher proportion of sons drank coffee on a daily basis
(25%) compared to sons of mothers with lower intake (16%). Results of sub-analyses for semen volume, sperm concentration,
and testosterone levels are reported to be unchanged when also
adjusted for the effects of maternal/son’s caffeine consumptions.
The authors acknowledge the third trimester assessment of
average exposure during pregnancy may not accurately reflect caffeine intake during the relevant window of testicular development.
Furthermore, exposure misclassification was likely given the
assignment of a mid-range caffeine concentration to all individuals
within a broad category of current coffee or cola intake. Like most
studies of semen quality, the participation rate was low (48.5%)
introducing the possibility of selection bias. Although this report
improves upon previous studies of semen quality, which had limited adjustment for confounders, the dichotomous measurement of
current and maternal smoking may have failed to fully control for
amount smoked. Overall, the study provides no indication that prenatal coffee consumption or current caffeine intake adversely affects semen quality in young adult males.
3. Spontaneous abortion
A good review of studies that have evaluated the relationship
between caffeine consumption and spontaneous abortion has been
published by Signorello and McLaughlin (2004).
3.1. Missing early losses
Some women conceive and then miscarry without any recognition of these events other than missing one period (Weinberg et al.,
1992; Wilcox et al., 1999). By all definitions, they had a spontaneous abortion, and yet they did not seek medical care and thus are
not included in studies of spontaneous abortion. Studies are biased
to the extent that they miss these early spontaneous abortions. The
best studies seek to improve outcome ascertainment by recruiting
women when they first become pregnant or even earlier when trying to conceive, utilizing biomedical advances in early pregnancy
detection. An alternative approach is to exclude all early losses before some specified gestational stage, e.g., 8 weeks, in an effort to
2556
create a homogeneous population for study. Because the causes of
early losses appear to be distinct from those of later miscarriages,
the results of such studies may not be readily extrapolated to populations at risk for early pregnancy loss.
Studies of caffeine and spontaneous abortion vary by the range
of gestational ages included for study (unspecified (Wen et al.,
2001; Giannelli et al., 2003); 6–12 weeks (Signorello et al., 2001),
<13 weeks (Maconochie et al., 2007); 6–16 weeks (Rasch, 2003);
620 weeks (Khoury et al., 2004; Weng et al., 2008; Savitz et al.,
2008b); <28 weeks (Tolstrup et al., 2003). Consequently, comparability is compromised.
3.2. Role of abnormal karyotypes
By and large, the earlier the miscarriage, the higher the proportion with chromosomal abnormalities (Yusuf and Naeem, 2004).
Until evidence is presented to the contrary, it seems highly desirable to evaluate the antecedents of karyotypically normal miscarriages separately from those of karyotypically abnormal
miscarriages. Only one study published since 2000 considered fetal
karyotype (Signorello et al., 2001).
3.3. Pregnancy signal
The potential for confounding introduced by the pregnancy signal, which was discussed earlier, is relevant to studies of caffeine
exposure and risk of spontaneous abortion. In this regard, smokers
are less likely to experience nausea and vomiting during pregnancy
than non-smokers (Weigel and Weigel, 1989; Louik et al., 2006).
Thus, the effects of confounding due to nausea may differ by smoking status, with greater impact on non-smokers whose reports of
increased caffeine intake may be more likely to reflect reduced
nausea that accompanies nonviable pregnancies.
3.4. Proper control selection in case-control studies
Most case-control studies of caffeine-miscarriage relationships
selected controls from among pregnant women seeking prenatal
care. This inclusion criterion is often not applied to the case group,
which includes women who had not sought routine prenatal care
before presenting with vaginal bleeding (Giannelli et al., 2003;
Rasch, 2003; Signorello et al., 2001; Karypidis et al., 2006; George
et al., 2006). If some controls receive advice from their prenatal
care provider to reduce caffeine intake, then selection bias will result. Some of this bias can be reduced if controls are interviewed
about their caffeine intake at the first prenatal care visit before
clinical advice would have an opportunity to influence intake patterns (Giannelli et al., 2003; Rasch, 2003).
3.5. The importance of pregnancy duration (gestational age)
Unless cases and controls are of comparable gestational age, the
exposure opportunity of one group will be larger than that of the
other. For example, the longer the duration of the pregnancy before
caffeine/coffee assessment, the greater the likelihood of a pregnancy signal and the opportunity to decrease consumption. Some
studies have selected controls from women who delivered a full
term live birth (Sata et al., 2005) or from women who have not
yet delivered (Giannelli et al., 2003), without consideration for
duration of pregnancy. The timing of control selection also has
implications for the quality of the retrospectively reported exposure data when the recall period is not similar for cases and controls. These errors can severely limit the value of a study’s findings.
3.6. Late recognition of fetal demise
Some fetuses die in utero weeks before being expelled (Simpson,
1990). Exposures during the time between fetal demise and recognition of the loss are irrelevant. To the extent that this time interval
is unrecognized, bias can result. Studies of caffeine and pregnancy
loss that attempt to exclude exposures occurring during the time
period immediately preceding recognition of the loss (Cnattingius
et al., 2000; Bech et al., 2005) may improve upon the approximation of the etiologically relevant time period.
3.7. Review of individual studies of spontaneous abortion
3.7.1. Cohort studies
Wen, W., Xiao, U.S., Jacobs, D.R., Brown, J.E., 2001. The associations
of maternal caffeine consumption and nausea with spontaneous abortion. Epidemiology 12, 38–42.
In this prospective study, women who consumed P100 mg/day
of caffeine during the first trimester were at elevated risk of spontaneous abortion [100–299 mg/day RR:2.0 (1.0–4.1); P300 mg/
day RR:2.5 (1.0–6.4)] compared to those consuming <20 mg/day
(n = 584). These increased risks were not apparent when the focus
was consumption prior to the onset of nausea (n = 206) or in women who never experienced nausea (n = 73), but were limited to
caffeine intake after the onset of nausea (n = 498) [RR:1.8 (0.8–
3.9), RR:2.4 (0.9–6.2) and RR:5.4 (2.0–14.6) for intakes of 20–99,
100–299 and P300 mg/day, respectively]. These findings could
be explained by the pregnancy signal.
Frequent assessments of caffeine intake throughout pregnancy
and evaluation of the interaction between caffeine and nausea
are strengths of this study. Limitations include the small number
of events in the stratified analyses which resulted in imprecise
measures of effect. In addition, control for potential confounders
may not have been sufficient, since they were ascertained only at
enrollment and were analyzed in broad categories (e.g., smoking
at enrollment, yes/no), creating the opportunity for residual confounding. Furthermore, assumptions were made about the temporal association of caffeine consumption and spontaneous abortion,
although the authors acknowledge that the exact timing of fetal
demise was unknown. Duration of nausea was explored by
restricting an analysis to women reporting nausea in at least two
monthly questionnaires, but results were imprecise and nonsignificant.
Hum. Reprod. 17, 1746–1754. [Discussed also in 2. Subfecundity
and 5. Gestational Age and Preterm Birth]
Miscarriage occurred in 21 of 62 confirmed pregnancies following in vitro fertilization (IVF) and gamete intra-fallopian transfer
(GIFT). Unstable but elevated point estimates were reported for
caffeine intake [OR:6.2 (0.9–40.8) for >50 mg/day] during the week
of the initial clinic visit. No associations with first trimester caffeine use were observed. The authors stated, ‘‘Because the sample
size was small, the CIs were very wide and the magnitude of association for caffeine and miscarriage may be unreliable”. The study
controlled for several potential confounders, but certain known
confounders such as smoking and pregnancy signal symptoms
were not controlled in the analysis. Thus, the results of this study
do not provide convincing evidence of a link between caffeine
and spontaneous abortion among women receiving assisted reproductive technology (ART) treatment.
Khoury, J.C., Miodovnik, M., Buncher, C.R., Kalkwarf, H., McElvy, S.,
Khoury, P.R., Sibai, B., 2004. Consequences of smoking and caffeine
consumption during pregnancy in women with type 1 diabetes.
J. Matern. Fetal Neonatal Med. 15, 44–50. [Discussed also in 5. Gestational Age and Preterm Birth and 6. Congenital Malformations]
This study of adverse pregnancy outcomes was conducted
among 191 pregnancies (23 spontaneous abortions 620 weeks)
identified within a cohort of women with type 1 diabetes who
were either pregnant or planning a pregnancy. Compared to no caffeine intake, the observed odds ratios for spontaneous abortions
620 weeks were 3.8 (0.8–16.9) for first trimester consumption of
1–2 cups of caffeinated beverages per day and 5.5 (1.2–22.0) for
P3 cups per day.
The major strength of this study is the prospective collection of
first trimester caffeine consumption, which avoids reporting bias
that can result from post-outcome assessments. The major limitations are lack of control for pregnancy symptoms, imprecise point
estimates due to small sample size and exposure misclassification
resulting from the less than optimal definition of caffeine consumption (monthly average number of 8 oz equivalent cups with
equal weighting for coffee, tea and soft drinks).
Bech, BH., Nohr, E.A., Vaeth, M., Henriksen, T.B., Olsen, J., 2005.
Coffee and fetal death: a cohort study with prospective data. Am.
J. Epidemiol. 162, 983–990. [Discussed also in 4. Fetal Death]
This prospective cohort study of spontaneous abortion
(<28 weeks) was conducted among 88,482 pregnancies within
the Danish National Birth Cohort. Women were enrolled following
their first prenatal visit and interviewed at 16 weeks of gestation
on average (inter-quartile range 13–19 weeks). The authors observed that heavy coffee drinkers (P8 cups per day) compared to
non-coffee drinkers had a higher risk of spontaneous abortion before 20 weeks of gestation [HR:1.5 (1.0–2.2)]. Despite the advantage of such large numbers, the limitations include missing early
spontaneous abortions, a lack of information about coffee consumption during the early weeks of pregnancy, and the absence
of attention to the implications of early pregnancy symptoms.
Weng, X., Odouli, R., Li, D.K., 2008. Maternal caffeine consumption
during pregnancy and the risk of miscarriage: a prospective cohort
study. Am. J. Obstet. Gynecol. 198, 279–286.
This cohort study enrolled 1063 health plan members after confirmation of pregnancy (median gestational age at interview was
10 weeks). Women reported average daily intake of caffeine-containing beverages since their last menstrual period and whether
their consumption patterns had changed since becoming pregnant.
Because 59% of miscarriages occurred prior to contact for the
study, caffeine consumption was reported retrospectively for a
substantial number of participants.
Compared to those with no caffeine exposure, the adjusted hazard ratio (HR) for miscarriage associated with caffeine intake
P200 mg/day was 2.2 (1.3–3.7). No significant interactions between consumption of caffeine P200 mg/day vs. <200 mg/day
and presence or absence of nausea or vomiting since LMP or smoking status were observed. Among women who reduced caffeine intake and were therefore most likely to have experienced strong
pregnancy signals, consumption of P200 mg/day was no longer
associated with miscarriage [HR:1.5 (0.9–2.5)]. Thus, the association with caffeine appears to be explained by the pregnancy signal.
Incomplete control for confounding by amount smoked or
severity or duration of nausea and vomiting was likely since these
factors were operationalized as dichotomous indicators (yes/no).
Caffeine intake measured through the end of pregnancy may also
have inadvertently captured consumption patterns that increased
following unrecognized fetal demise. Despite concerns about potential recall bias, results were similar when analyses were restricted to women interviewed before the miscarriage occurred
(aHR = 2.8; 95% CI 1.1–7.0) (Li et al., 2008). However, confounding
by the pregnancy signal is reasonably supported.
Savitz, D.A., Chan, R.L., Herring, A.H., Howards, P.P., Hartmann, K.E.,
2008. Caffeine and miscarriage risk. Epidemiology 19, 55–62.
2557
This study evaluated 258 spontaneous abortions (620 weeks of
gestation) occurring among a cohort of 2407 women with clinically
recognized pregnancies. Caffeine assessment considered daily consumption in a ‘typical week’ during three time periods – pre-pregnancy, four weeks after last menstrual period (LMP), and at the
time of interview, referred to as ‘‘current consumption” (i.e., before
16 completed weeks or when still pregnant for those with losses).
Overall, the study reported no association between coffee consumption or total caffeine intake before or during pregnancy and
risk of spontaneous abortion up to 20 weeks of gestation.
Among women interviewed after the miscarriage, current caffeine consumption P144.3 mg/day was associated with an increased risk of spontaneous abortion [OR:1.9 (1.1–3.5)] compared
to non-consumers. In contrast, among women interviewed before
their loss, no association with current caffeine intake was observed
[OR:1.1 (0.6–1.8)]. These disparate results could reflect recall bias,
increased consumption patterns following fetal demise, true differences in consumption due to absence of a strong pregnancy signal,
a relationship between caffeine consumption and later miscarriages only, or chance.
This study has several strengths including a large sample, a prospective design, use of an analytic approach that estimates the
probability of having a loss in a given week, conditional on still
being pregnant (i.e., at risk of spontaneous abortion) at the beginning of that week, and use of median and 75th percentile cutpoints for defining exposure categories, which increase precision
when biologically meaningful cut-points are unknown.
The results of this study do not support an association between
spontaneous abortion and caffeine intake before or during pregnancy at the levels assessed. The most likely explanation of the
associations produced by the sub-analysis of exposure data collected after the event of a loss is recall bias.
3.7.2. Nested case-control studies
Tolstrup, J.S., Kjaer, S.K., Munk, C., Madsen, L.B., Ottesen, B., Bergholt, T., Grønbaek, M., 2003. Does caffeine and alcohol intake before
pregnancy predict the occurrence of spontaneous abortion? Hum.
Reprod. 18, 2704–2710.
This study used an atypical definition of spontaneous abortion
as pregnancy loss within the first 28 weeks of pregnancy, thus
complicating comparisons with other studies which apply the clinical definition, loss before 20 weeks. This nested case-control study
was conducted within a cohort of 11,088 non-pregnant women
randomly sampled from the general population of 20–29 year olds
living in Copenhagen, Denmark. A baseline interview was conducted between May 1991 and January 1993. A follow-up interview was administered 2 years later to ascertain all pregnancies
(n = 1381) and related outcomes (n = 303 spontaneous abortions)
occurring during the follow-up interval. The Danish Hospital Discharge Registry was also used to confirm interview reports and
to ascertain pregnancy outcomes in women who did not complete
the follow-up questionnaire.
The adjusted odds ratio of miscarriage before 28 weeks gestation was significantly increased only for women consuming
>900 mg caffeine per day from coffee or tea [OR:1.7 (1.0–3.0)],
after adjustment for maternal age, marital status, cigarette smoking and alcohol intake. At consumption levels similar to those
examined by others (e.g., 300 mg/d), average daily pre-pregnancy
intake of caffeine was not associated with miscarriage.
This study had a number of methodological features that limit
confidence in the results. Because of the time lag between ascertainment of caffeine exposure from the first interview and conception (mean 9.3 ± 6.5 months), caffeine data obtained at baseline
may misclassify patterns of consumption more proximal to conception. Pregnancy symptoms were also not assessed and thus
were not controlled in the analysis. Furthermore, while it is possi-
2558
ble that the 20% of self-reported miscarriages that were not registry-confirmed represented very early fetal losses among women
who did not seek medical care, it is also possible that they were
not miscarriages. Test results to confirm pregnancies were not
available.
3.7.3. Case-control studies
Giannelli, M., Doyle, P., Roman, E., Pelerin, M., Hermon, C., 2003.
The effect of caffeine consumption and nausea on the risk of miscarriage. Paediatr. Perinat. Epidemiol. 17, 316–323.
This case-control study was conducted among nulliparous women seeking care for spontaneous abortion (n = 160 cases) or pregnancy (n = 314 controls). Cases were interviewed about 3 weeks
after their pregnancy loss on average, whereas controls were interviewed at the first prenatal care visit which typically occurred at a
more advanced gestational age.
When compared to women reporting <151 mg/day, women
consuming more than 300 mg/day during pregnancy were at increased odds for spontaneous abortion [OR:1.9 (1.0–3.6) for 301–
500 mg/day and OR:2.2 (1.1–4.4) for P500 mg/day]. No associations were observed for pre-pregnancy caffeine use.
The burden of recalling caffeine exposure was not equivalent for
cases and controls. The authors report collecting data on the timing
and explanation for variations in typical caffeine consumption patterns before and during pregnancy (but not actual change in
amount consumed), although it does not appear that such changes
were incorporated into the exposure estimates.
A strength of this study is its attempt to control for severity of
nausea. As anticipated, significantly fewer cases than controls reported nausea in this study population (45% of case and 81% of controls interviewed in first trimester) and spontaneous abortion was
strongly associated with less severe nausea. Despite these strong
associations, controlling for severity of nausea (none, mild, moderate, and severe) had little impact on the magnitude of the reported
point estimates and no interaction between nausea and caffeine intake was observed.
The limitations of this study include potential recall bias,
incomplete control for smoking (excluded as a potential confounder when measured as yes/no) and inclusion of caffeine intake
following fetal demise. The 2-fold increased odds observed in this
study are similar in magnitude to the association produced in
Savitz et al. (2008b) when exposure data were retrospectively
reported by cases as compared to prospective reporting. The study
also did not consider dietary aversions as an indicator of the pregnancy signal phenomenon, as sensitivity to odors and/or food/beverage aversions may be more closely associated with decreased
caffeine consumption than nausea severity.
Rasch, V., 2003. Cigarette, alcohol, and caffeine consumption: risk
factors for spontaneous abortion. Acta Obstet. Gynecol. Scand. 82,
182–188.
This case-control study included 303 women with clinically recognized spontaneous abortions between 6 and 16 weeks of gestation and 1168 control women pregnant with a live fetus in the
same range of gestational age. Women reporting heavy caffeine
use since becoming pregnant (P375 mg/day) had a 2.2 (1.5–3.2)
times greater odds of spontaneous abortion compared to women
consuming 0–199 mg/day.
The major strengths of this study include the use of control
pregnancies of similar gestational ages, which appropriately represents the source population for the cases, the relatively large sample size, and the high proportion of heavy caffeine users in the
study population (46.6% of cases and 28.9% of controls). Among
the study limitations considered by the author are no data to control for pregnancy symptoms, the possibility of including exposure
following fetal demise and the possibility of overestimation due to
recall bias given cases reported exposures while being hospitalized
for evacuation of the pregnancy. The study author reasons recall
bias is an unlikely explanation because concern about caffeine
use during pregnancy was not widespread in Denmark during
the mid-1990’s study period.
A substantial proportion of the study population was excluded
due to missing data on gestational age (23.8% of case respondents
and 21.4% of controls). In addition, it is not possible to disentangle
the evidence to determine whether the observed associations reflect something other than a diminished or absent pregnancy signal in pregnancies destined to fail.
Maconochie, N., Doyle, P., Prior, S., Simmons, R., 2007. Risk factors
for first trimester miscarriage – results from a UK-population-based
case-control study. Br. J. Obstet. Gynaecol. 114, 170–186.
Data for this case-control study were derived from the National
Women’s Health Study of adult women living in the United Kingdom in 2001. Women with a pregnancy history identified in a
two-stage population-based survey were selected as cases if their
last pregnancy ended in a first trimester miscarriage (<13 completed weeks) or if they had a miscarriage in any pregnancy conceived since 1995 (n = 603). This latter group represented about
40% of all cases. Controls were women whose last pregnancy progressed to 13 weeks gestation or beyond (n = 6116). Eighty-three
percent of cases were conceived since 1995, compared to 49% of
controls, thus there was potential for differential recall between
cases and controls.
Average caffeine intake from beverages >300 mg/day during the
first 12 weeks of pregnancy was independently associated with an
increased odds of miscarriage compared to no consumption
[OR:1.5 (1.1–2.2) for 301–500 mg/d; OR:1.7 (1.2–2.4) for
>500 mg/d]. After further adjustment for nausea severity, however,
caffeine intake was no longer related to the odds of miscarriage
[OR:1.0 (0.7–1.5) for 301–500 mg/d and OR:1.1 (0.8–1.7) for
>500 mg/d].
The limitations of this study include the potential misclassification of exposure and outcome due to lengthy and differential
time frames for exposure recall, self-reported outcomes, and confining caffeine intake to beverages. These data do not support an
association between caffeine consumption from beverages and
odds of first trimester miscarriage after adjusting for the confounding effect of nausea severity.
3.7.4. Caffeine metabolism
Signorello, L.B., Nordmark, A., Granath, F., Blot, W.J., McLaughlin,
J.K., Annerén, G., Lundgren, S., Ekbom, A., Rane, A., Cnattingius, S.,
2001. Caffeine metabolism and the risk of spontaneous abortion of
normal karyotype fetuses. Obstet. Gynecol. 98, 1059–1066.
This study is one of three papers (Signorello et al., 2001; Karypidis et al., 2006; George et al., 2006) published using data collected
as part of a case-control study of caffeine and miscarriage in Uppsala, Sweden (Cnattingius et al., 2000). The original publication by
Cnattingius et al., was previously reviewed by Leviton and Cowan
(2002). In the original study, 562 confirmed cases of spontaneous
abortions between 6 and 12 weeks of gestation were identified
from a university hospital, the sole source of care for women experiencing pregnancy loss. Controls (n = 953) were identified from
women seeking prenatal care in Uppsala and were frequency
matched to cases by completed weeks of gestation and area of residence. This study was well-designed in that controls represented
the source population of the cases, i.e., women in their first trimester of pregnancy. A major strength of this study was karyotyping of
cases when fetal tissue could be obtained, although this was available on fewer than half of the cases. Another positive quality was
the measurement of caffeine consumption, which although ascertained retrospectively, incorporated multiple sources of exposure,
method of preparation, serving size, and captured reports of caffeine use week by week. The potential drawback of the exposure
assessment was the likely inclusion of caffeine intake following
unrecognized fetal demise.
The 101 chromosomally normal spontaneous abortions were
compared to the 953 controls. With the goal of evaluating variability in caffeine metabolism as a risk factor for spontaneous abortions, the authors estimated the activity levels of two enzymes,
cytochrome P4501A2 (CYP1A2) and N-acetyltransferase 2 (NAT2),
given both are involved in the metabolism or detoxification of caffeine (Campbell et al., 1987a; Nebert et al., 1996; Arnaud, 1994).
Using genomic DNA from whole blood, polymorphisms of the
NAT2 gene were determined by polymerase chain reaction (PCR)
amplification to identify slow (homozygous mutated alleles) and
fast acetylator genotypes (heterozygous or homozygous wild type
alleles). CYP1A2 phenotypes were determined using the index described by Campbell et al. (1987b), which calculates a ratio of four
urinary metabolites of caffeine as an indicator of caffeine clearance.
Low and high CYP1A2 activity were defined as log-transformed
CYP1A2 index values below and above the median value identified
among controls (median index value = 0.73).
When stratified by CYP1A2 activity, caffeine intake above
100 mg/day was associated with increased odds of spontaneous
abortion among women with high CYP1A2 activity [OR:2.4 (1.0–
5.8) for 100–299 mg/day and OR:3.2 (1.2–8.2) for P300 mg/day
compared to 0–99 mg/day], but not among those with low activity.
Independent of caffeine intake, smoking status, nausea score,
maternal age and week of gestation, high CYP1A2 activity was
associated with an elevated odds of spontaneous abortion
[OR:2.9 (1.7–5.0)]. NAT2 genotype was not associated with spontaneous abortion. The authors suggest that prolonged exposure to
caffeine metabolites might explain their unexpected findings.
A limitation of using spot urine samples in population-based
studies to phenotype CYP1A2 activity (as opposed to the less feasible method of administering a standardized test dose of caffeine) is
that recent caffeine consumption is required in order to detect the
caffeine metabolites in urine (Nordmark et al., 1999). Thus, women
consuming little or no caffeine (approximately 1/3 of cases and
controls) were excluded from the analyses. Furthermore, the exclusions resulted in small sample sizes and imprecise estimates, particularly within the stratified analyses.
Because CYP1A2 is involved in the metabolism of numerous
drugs in addition to caffeine, the authors recognize that the results
may indicate potential links between spontaneous abortion and
other unmeasured factors. Cigarette smoke is one compound that
has been shown to induce high CYP1A2 activity (Campbell et al.,
1987b). Classifying women as either smokers or non-smokers, instead of actual quantity smoked (or concentration of plasma cotinine) might have contributed to residual confounding by
smoking. This may also explain why high CYP1A2 activity was
associated with spontaneous abortion contrary to expectations
that risk would increase with slower caffeine clearance. Thus, caution is advised when drawing inferences from these findings.
Karypidis, A.H., Söderström, T., Nordmark, A., Granath, F., Cnattingius, S., Rane, A., 2006. Association of cytochrome P450 1B1 polymorphism with first-trimester miscarriage. Fertil. Steril. 86, 1498–1503.
The Swedish case-control study was utilized again by Karypidis
et al. (2006) to evaluate the odds of spontaneous abortion associated with CYP1B1 polymorphisms and a possible interaction with
caffeine consumption. CYP1B1 is an enzyme that influences the
metabolism of steroid hormones, such as testosterone, progesterone, and estradiol, and of caffeine, among other agents. CYP1B1
activity, like CYP1A2, is also induced by smoking (Piipari et al.,
2000). Women who were homozygous for the Val allele had an adjusted odds of miscarriage 1.5 times higher than women homozygous for the Leu allele (95% CI = 1.0–2.1), and the association was
consistent among non-smokers [OR:1.6 (1.1–2.4)]. The risk increased with increasing caffeine consumption, but almost only
2559
among Val/Val. This indicates a significant interaction between
homozygosity for Val and caffeine intake.
Since the Leu variant contributes to inactivation of testosterone
(Shimada et al., 1999) and higher levels of testosterone have been
associated with miscarriage (Okon et al., 1998), Leu/Leu homozygotes may be protected from spontaneous abortion. The authors
acknowledge that CYP1B1 is involved in the metabolism of many
compounds including steroid hormones and procarcinogens as
well as caffeine; thus, the observed associations may not be directly attributed to caffeine metabolism.
3.7.5. Recurrent pregnancy loss
George, L., Granath, F., Johansson, A.L.V., Olander, B., Cnattingius, S.,
2006. Risks of repeated miscarriage. Paediat. Perinat. Epidemiol. 20,
119–126.
This case-control study of repeated miscarriage was conducted
within the Swedish study of spontaneous abortions reported by
Cnattingius et al. (2000). Cases included 108 women from the original case group who had two or more consecutive miscarriages.
Controls were women with at least two pregnancies (n = 583), of
which the last was a normal intrauterine pregnancy, confirmed
by vaginal ultrasound.
After adjustment for potential confounders, the odds of
repeated miscarriage was not significantly increased in heavy
caffeine users [OR:1.8 (0.8–3.9) for P300 mg/day]. Caffeine consumption P300 mg/day was related to repeated miscarriage in
non-smokers [OR:2.7 (1.1–6.2)] but not smokers [OR:0.4 (0.05–
4.1)]. The test for interaction was not statistically significant.
As with the original study, the increased odds of repeated miscarriage associated with high caffeine intake only in non-smokers
may be a result of lower prevalence of other risk factors for miscarriage in non-smokers, making a relation with caffeine more apparent. It may also be due to the fact that smoking, as the authors
report, increases the rate of elimination of caffeine. Lack of control
for pregnancy symptoms could provide an alternative explanation
for the association between caffeine consumption and repeated
spontaneous abortion in non-smokers, as described in Section 3.3.
3.7.6. Recurrent pregnancy loss and caffeine metabolism
Zusterzeel, P.L., Nelen, W.L., Roelofs, H.M., Peters, W.H., Blom, H.J.,
Steegers, E.A., 2000. Polymorphisms in biotransformation enzymes
and the risk for recurrent early pregnancy loss. Mol. Hum. Reprod. 6,
6474–6478.
This case-control study of recurrent early pregnancy loss evaluated associations with polymorphisms in glutathione S-transferase
(GST) and cytochrome P450 genes. The authors reasoned that polymorphisms in these genes may reflect impaired drug metabolism
and detoxification, which could increase susceptibility to adverse
outcomes resulting from chemical exposures. The study included
187 case women who had at least two unexplained consecutive
spontaneous abortions occurring <17 weeks of gestation. The cases
were selected from a referral hospital and 109 controls were selected from unrelated acquaintances who had at least one uncomplicated pregnancy and no spontaneous abortions (and were, thus,
considered matched by age, socioeconomic status and district). The
ratio of less than one control per case is not explained, but suggests
differential response rates for cases and controls. Details regarding
coffee exposure assessment are also omitted.
Odds ratios for the effects of caffeine consumption were not
reported, but our calculations show no observed associations
between daily coffee intake and recurrent pregnancy loss [crude
OR:0.8 (0.4–1.5) for 1–5 cups and crude OR:1.4 (0.7–2.9) for
>(and equal to ?) five cups compared to non-coffee drinkers]. The
non-mutually exclusive categories of coffee consumption are those
reported by the authors.
2560
GSTP1b-1b genotype (representing lower enzyme activity) was
associated with recurrent pregnancy loss (OR:2.9 (1.0–10.1), especially among coffee drinkers (OR:4.1 (1.2–13.3). The increased OR
among coffee drinkers could be attributable to poorer precision
in the smaller subgroup, which is based on only three controls
homozygous for the GSTP1b allele. It could also be explained by
uncontrolled confounding by smoking.
Although the GSTP1b–1b polymorphism appeared to be more
common among women with recurrent early pregnancy loss, the
limited data presented in this paper offer little evidence to implicate a specific role for coffee intake via direct effects or interactive
effects with GST polymorphisms.
Sata, F., Yamada, H., Suzuki, K., Saijo, Y., Kato, E.H., Morikawa, M.,
Minakami, H., Kishi, R., 2005. Caffeine intake, CYP1A2 polymorphism
and the risk of recurrent pregnancy loss. Mol. Hum. Reprod. 11, 357–360.
The authors of this small case-control study (58 cases and 147
controls) reported no overall association between caffeine intake
P300 mg/day and recurrent pregnancy loss [OR:1.8 (0.7–4.6)]
when compared to intake of 0–99 mg/day. Among women homozygous for CYP1A2 1F (A/A) alleles, considered to have heightened caffeine metabolism, caffeine intake P300 mg/day was
associated with recurrent pregnancy loss [OR:5.2 (1.1–25.9)] when
compared to limited caffeine use (0–99 mg/day). Effect modification by CYP1A2 genotype was not observed among women with
other CYP1A2 genotypes.
The limitations of this study include the small sample size, poor
response rates (56.6% for cases and 58.1% for controls), lack of
information on the timing of recruitment in relation to the previous pregnancy and implications for accurate exposure recall, no
assessment of pregnancy symptoms, and the possibility of residual
confounding by smoking which was measured as never, quitter
and continuous smoker. Smokers homozygous for the CYP1A2 F1
(A/A) allele have higher CYP1A2 activity compared to other
CYP1A2 genotypes (A/C and C/C) (Sachse et al., 1999). Thus, associations within the CYP1A2 F1 strata may be explained by smoking,
particularly if reported caffeine use reflects actual smoking patterns (as described by Morrison, 1984).
The methodological limitations do not allow unambiguous conclusions to be drawn about interactive effects of heavy caffeine use
and CYP1A2 genotype on recurrent pregnancy loss.
4. Fetal death
Clinical convention distinguishes fetal death, defined as fetal
demise after 20 weeks of gestation, from spontaneous abortion defined as pregnancy loss 620 weeks of gestation. The distinction is
typically drawn at the mid-point of pregnancy because it approximates the point of fetal viability, which is generally regarded as
occurring close to 23–24 weeks of gestation. Because both outcomes address fetal loss, but at different points along the continuum of pregnancy, many of the methodological considerations
identified for studies of spontaneous abortions are also applicable
to studies of fetal death.
Studies of caffeine and fetal death tend to focus on mid-pregnancy exposures and thus may be particularly subject to confounding by the pregnancy signal. Most women decrease their coffee
consumption during the early part of pregnancy and maintain their
lower than pre-pregnancy levels of consumption throughout the
remainder of their pregnancy (Boylan et al., 2008). Thus, women
who are heavy coffee consumers in the last half of pregnancy
may not have experienced a strong pregnancy signal early in pregnancy. Only one study of fetal death (Matijasevich et al., 2006) con-
sidered nausea and vomiting, but measurement was limited and
food/drink aversions were not considered.
4.2. Heterogeneity
Efforts have been made to group fetal deaths into more etiologically homogeneous outcomes according to the time of fetal demise (antepartum vs. intrapartum; early fetal deaths <28 weeks
vs. stillbirths P28 weeks) and cause of death. The evaluation of fetal deaths by likely causes (e.g., congenital infection, malformations, placental abruption) is hindered by the difficulty of
attributing deaths to a single cause, the challenge of systematic
identification of all contributing causes, and the need for large
sample sizes (Leviton, 1987). While a worthwhile goal, the quality
of analyses which classify outcomes by cause of fetal death remain
questionable.
4.3. Selection of controls
Controls selected from live, healthy births (Matijasevich et al.,
2006; Bech et al., 2006) may introduce selection bias. As with studies of spontaneous abortion, controls selected from on-going pregnancies matched by gestational age would provide the most
accurate representation of the exposure distribution among the
population that generated the cases of fetal deaths.
4.4. Late recognition of fetal demise
The exact timing of death is rarely observed, but can occur
weeks before the fetal loss is recognized. Like studies of spontaneous abortion, the inclusion of caffeine intake up to the point of
pregnancy termination may inflate associations with fetal deaths
if the diminished pregnancy signal following fetal demise leads
to increased caffeine consumption (Stein and Susser, 1991).
4.5. Review of individual studies of fetal deaths
Wisborg, K., Kesmodel, U., Bech, B.H., Hedegaard, M., Henriksen,
T.B., 2003. Maternal consumption of coffee during pregnancy and stillbirth and infant death in first year of life: prospective study. BMJ 326,
420–423.
This prospective cohort study of 18,478 singleton pregnancies
in Denmark reported an increased odds of stillbirth (P28 completed weeks of gestation) among pregnant women drinking P8
cups of coffee per day compared to non-coffee drinkers [OR:2.2
(1.0–4.7]. The strength of this study was its prospective design,
which avoided opportunities for recall bias. The measurement of
caffeine/coffee intake, however, was limited to current coffee consumption at 16 weeks of gestation, without reference to a standard
cup size. Although other sources of caffeine intake were collected,
the authors ignored contributions from tea, cola or chocolate since
few women reported high intakes from these sources. Because
3922 women contributed more than one pregnancy to the cohort,
sub-analyses addressing the lack of independence between observations were conducted. Results were not presented, but were reported as ‘‘comparable”. If women with less viable pregnancies
consume more caffeine because they have fewer pregnancy symptoms and aversions, then higher caffeine consumption would be a
consequence of an unhealthy pregnancy rather than a cause. Wisborg et al. (2003) did not consider pregnancy signal symptoms in
their analyses.
Bech, B.H., Nohr, E.A., Vaeth, M., Henriksen, T.B., Olsen, J., 2005. Coffee and fetal death: a cohort study with prospective data. Am. J. Epidemiol. 162, 983–990. [Discussed also in 3. Spontaneous Abortion]
In this prospective study of 88,482 pregnancies within the Danish National Birth Cohort, the fetuses of women who consumed 8
or more cups of coffee per day were at increased risk of death between 20 and 27 weeks of gestation [HR:2.3 (1.3–3.9)], but were
not at increased risk of death after 27 weeks [HR:1.3 (0.7–2.4)].
Risk of stillbirth due to placental dysfunction was 2.3 times
greater among women consuming P4 cups of coffee per day compared to women who consumed no coffee [HR:2.3 (1.2–4.3)]. No
associations were observed for other stillbirth subgroups such as
unexplained intrauterine deaths, umbilical cord complications,
congenital malformations, ‘‘other” conditions such as infection
and maternal disease, or intrapartum deaths.
To minimize bias that can occur when caffeine consumption is
increased as a result of undetected fetal demise, the authors repeated analyses after excluding losses occurring over a range of
2–28 days following the interview. The hazard ratios became
attenuated with increasing lag time and non-significant in all
exposure categories, suggesting the results could be explained by
the pregnancy signal. Although the data are not shown, the authors
note the decline in hazard ratios to be particularly strong among
fetal deaths <20 weeks, but less so for fetal deaths P20 weeks.
Associations with later fetal losses may be less susceptible to the
influence of undetected fetal demise since exposure assessment
(between 13 and 19 weeks of gestation) preceded the loss by a
longer time period. However, reverse causation remains a possibility if higher coffee intake is a marker of placental dysfunction and
thus a consequence of the reduced or absent symptoms that
accompany less viable pregnancies, regardless of the timing of fetal
demise. The authors acknowledge their inability to assess the impact of pregnancy symptoms on the observed associations.
Bech, B.H., Autrup, H., Nohr, E.A., Henriksen, T.B., Olsen, J., 2006.
Stillbirth and slow metabolizers of caffeine: comparison by genotypes.
Int. J. Epidemiol. 35, 948–953.
In this paper, the same research group as above conducted a
nested case-control study within the Danish National Birth Cohort
to evaluate the interrelationship between caffeine, caffeine metabolism and stillbirth. Cases (n = 142) were defined as stillbirths
occurring P28 weeks of gestation, excluding intrapartum deaths.
Controls (n = 157) were selected from live births, frequency
matched by parity. The authors evaluated genotypes either known
[cytochrome P4501A2 (CYP1A2) and N-acetyltransferase 2
(NAT2)], or suspected [glutathione S-transferase a1 (GSTA1)] to
be active in caffeine metabolism. CYP1A2 genotypes were grouped
into fast (A/A) and slow (A/C and C/C) oxidizers. NAT2 genotypes
were grouped into fast (Fast/Fast and Fast/Slow) and slow (Slow/
Slow) acetylators. GSTA1 genotypes were classified according to
high activity (a/a) and reduced activity (a/b and b/b). Coffee was
the only source of caffeine considered in this analysis. The methods
used to ascertain coffee intake are not provided, but the limitations
are presumed to be the same as described for Bech et al. (2005)
(see Section 3.7.1).
When caffeine use was assessed without consideration for metabolic genotype, drinking P4 cups of coffee per day was not associated with stillbirth [OR:1.0 (0.5–2.3)]. Women with stillbirths
had a 1.9-fold increased odds of having a slow caffeine metabolism
as characterized by all three genotypes (slow CYP1A2, slow NAT2
and low GTSA1) compared to controls [OR:1.9 (1.0–3.4)]. However,
when genotypes were assessed separately or in paired combinations, no associations with stillbirth were observed. The lack of
interaction between genotype and caffeine consumption suggests
the associations with these genotypes may not reflect causal pathways specific to caffeine metabolism.
Matijasevich, A., Barros, F.C., Santos, I.S., Yemini, A., 2006. Maternal
caffeine consumption and fetal death: a case-control study in Uruguay.
Paediat. Perinat. Epidemiol. 20, 100–109.
Cases (n = 382) included all antepartum fetal deaths
(P20 weeks of gestational age or weighing >350 g) occurring in
the 16 maternity hospitals within the capital city of Uruguay. Con-
2561
trols (n = 792) were healthy, full term, live births without growth
restriction, frequency matched by hospital. Participants were interviewed within 24 h following delivery for caffeine intake from
matè (herbal tea) and coffee during each trimester. The justification for limiting assessment to coffee and maté was that these
are the primary sources of caffeine in South America. However,
according to their own report of caffeine consumption within the
South American region (Santos et al., 1998), up to 48% of pregnant
women reported consumption of other sources such as soft drinks,
chocolate bars, and black tea, which accounted for approximately
one-fourth of their total caffeine intake during pregnancy. Thus,
exposure misclassification is likely.
The authors report that mean caffeine consumption of
P300 mg/day throughout pregnancy was associated with fetal
death [OR:2.3 (1.2–4.4)]. Several maternal characteristics were
considered as confounders, although the criteria for confounding
relied on p values, which can result in an appreciable loss of information (Dales and Ury, 1978). The crude measurement of smoking
(yes/no) and pregnancy symptoms (yes/no for vomiting or nausea
in first trimester) and failure to consider alcohol intake may have
also contributed to incomplete control of confounding.
As noted by the authors, recall bias could have inflated the association if case mothers reported past caffeine use more accurately
than controls. Another limitation of this study was the use of live
births as controls, which may not accurately represent the exposure distribution in the population from which the cases arose
(Signorello and McLaughlin, 2004).
In light of these limitations, this study does not provide convincing evidence that high caffeine consumption throughout pregnancy is associated with fetal death P20 weeks of gestation.
5. Gestational age and preterm birth
5.1. Heterogeneity
Preterm birth includes a heterogeneous group of disorders,
suggesting heterogeneous etiologies (Savitz et al., 1991, 2005;
Klebanoff and Shiono, 1995; Klebanoff, 1998a; Pennell et al.,
2007; Berhman and Butler, 2007; McElrath et al., 2008; Savitz,
2008a). The so-called ‘‘spontaneous preterm delivery” group includes preterm labor, pre-labor premature rupture of membranes,
placental abruption, and cervical insufficiency, which are associated with intrauterine inflammation. Medically indicated preterm
births can be characterized by maternal and fetal origins. The
maternal medical indications group is almost exclusively preeclampsia, and is attributed to dysfunctional placentation. Fetal
indications are heterogeneous and include non-reassuring fetal
testing, oligohydramnios, Doppler abnormalities of umbilical cord
blood flow, or severe intrauterine growth restriction identified on
antepartum ultrasound examination. Only two studies attempted
to evaluate relatively homogenous outcomes (Mikkelsen et al.,
2008; Haugen et al., 2008).
It is not yet clear that the processes leading to delivery near the
boundary of viability (23–24 weeks) are the same as those that
lead to delivery near the upper boundary of prematurity (34–
36 weeks). The two studies that acknowledged the possibility that
early and late prematurity might be different entities dichotomized premature deliveries at 35 weeks, rather than a considerably earlier week in pregnancy (Mikkelsen et al., 2008; Haugen
et al., 2008).
5.2. Accurate estimation of gestational age
Correct classification of preterm birth is dependent on accurate
estimates of gestational age. The three most commonly used meth-
2562
ods are based on ultrasound, date of last menstrual period and neonatal assessment of physical and neurological maturity at birth
(Lynch and Zhang, 2007). No method is perfect, but the use of early
ultrasound is considered to provide the most accurate estimation,
within 5–10 days when completed at 12–14 weeks of gestation
and 9–12 days when completed at 15–20 weeks of gestation
(Saltvedt et al., 2004).
Gestational age based on date of last menstrual period can be
flawed due to recall errors or delayed ovulation, commonly overestimating pregnancy duration (Savitz et al., 2002). In the absence of
other information, gestational age is sometimes estimated using
neonatal evaluations such as the Dubowitz and Ballard examinations, which score the physical and neuromuscular development
of the newborn (Dubowitz et al., 1970; Ballard et al., 1979,
1991). These postnatal estimates are less precise and accurate than
the preferred prenatal methods and have a tendency to underestimate gestation for post-term deliveries and to overestimate gestation by up to 2 weeks for deliveries occurring before 40 weeks
(reviewed by Lynch and Zhang, 2007).
Considering the deficiencies in each of these methods, some
have recommended a hierarchy of assessments ordered by accuracy and availability of the measurements. One example preferred
the dates of embryo retrieval or intrauterine insemination or ultrasound examination before the 14th week of gestation. If these were
not available, the order of preference continued with ultrasound at
14 weeks or later, followed by menstrual dating without ultrasound confirmation, and finally gestational age recorded after neonatal assessment (McElrath et al., 2008).
5.3. Review of individual studies of preterm birth
Clausson, B., Granath, F., Ekbom, A., Lundgren, S., Nordmark, A.,
Signorello, L.B., Cnattingius, S., 2002. Effect of caffeine exposure during
pregnancy on birth weight and gestational age. Am. J. Epidemiol. 155,
429–436. [Discussed also in 7. Fetal Growth]
In this cohort study from Sweden, 873 pregnant women were
interviewed twice, between gestational weeks 6–12 and 32–34,
for their recall of caffeine consumption during the previous weeks
of pregnancy. Although gestational age was confirmed by ultrasound examination, spontaneous and medically indicated preterm
births were not considered separately. Combining all preterm
births <37 weeks of gestation could potentially obscure existing
associations with caffeine intake. According to the authors, neither
mean length of gestation nor mean birth weight differed for mothers grouped according to average daily caffeine intakes across the
entire pregnancy (4–34 weeks of gestation) of 0–99, 100–299,
300–499 or P500 mg/day. Furthermore, average daily caffeine intake in each trimester was not associated with duration of
pregnancy.
Retrospective exposure assessment may have resulted in errors
in caffeine measurement given the difficulty of accurately remembering intake patterns across previous weeks and months. Since
consumption patterns were assessed before delivery, however, it
is unlikely that women with shorter gestations would have been
influenced to systematically report more or less caffeine consumption. The authors controlled for smoking by using third trimester
plasma cotinine concentrations. Pregnancy symptoms were also
assessed as confounders, but had little impact on the results. These
data do not support an association between caffeine and preterm
birth, however caffeine exposures after 32–34 weeks could not
be evaluated.
Klebanoff, M.A., Levine, R.J., Clemens, J.D., Wilkins, D.G., 2002.
Maternal serum caffeine metabolites and small-for-gestational age
birth. Am. J. Epidemiol. 155, 32–37. [Discussed also in 7. Fetal
Growth]
This study measured paraxanthine, the major metabolite of caffeine, in third trimester serum samples banked for 2515 women
participating in the Collaborative Perinatal Project (CPP) between
1959 and 1966. The subjects in this study served as the controls
in a previous case-control study of spontaneous abortion (Klebanoff et al., 1999); thus, all had delivered liveborn infants
P28 weeks of gestation. Serum samples collected after 26 weeks
of gestation were stored frozen for over 30 years before caffeine
metabolite concentrations were measured. After controlling for
maternal ethnicity, paraxanthine concentrations were not associated with gestational age or preterm birth.
The stability of caffeine metabolites in serum is unknown. In response to concerns about measurement error raised by Grosso
et al. (2004), the authors responded by demonstrating that detected concentrations of serum paraxanthine following long-term
storage were consistent with self-reported intake in another cohort
assembled during the same time period (Klebanoff and Longnecker, 2004).
Because the CPP study was conducted in the early 1960s, caffeine use during pregnancy was relatively common in the study
population, although actual consumption patterns were not directly assessed. Other strengths of the study include the opportunity to assess a biomarker of caffeine exposure that avoids
measurement errors related to inaccurate recall and the difficulty
of accounting for all sources. In light of individual differences in
caffeine metabolism, paraxanthine levels may reflect biologic dose
better than reported caffeine consumption. On the other hand, the
paraxanthine biomarker represents recent caffeine intake (halflife = 10 h in later pregnancy) (Aldridge et al., 1981) and may reflect exposure during the relevant window of susceptibility only
for those whose intake patterns remain relatively constant over
time.
The data do not support an association between third trimester
paraxanthine concentrations and pregnancy duration or preterm
delivery.
Hum. Reprod. 17, 1746–1754. [Discussed also in 2. Subfecundity
and 3. Spontaneous Abortion]
Within this study of couples undergoing IVF and GIFT
(described in Section 2.6), the association between caffeine intake and gestational age was assessed in 39 live births. No association was observed between caffeine intake among the male
partners and gestational age. Maternal caffeine intake >50 mg/
day during the week of the first fertility clinic visit was associated with a 3.5 week decrease (95% CI 6.7, 0.3) in gestational
age when compared to women reporting 0–2 mg/day. Similar
results were reported for usual ‘‘lifetime” caffeine consumption,
but results for intake during pregnancy were not reported. The
authors cautiously report the findings as having ‘‘borderline
significance” while acknowledging the small sample size and
limited precision.
Although the results were reported as adjusted, the specific
control variables were not specified. Factors controlled in analyses
of other outcomes in this report included smoking, alcohol use,
woman’s age, race, years of schooling, parity, type of infertility,
type of procedure, and number of ART attempts. However, with
only 39 observations, the precision of the results would be questionable with this many covariates (Feinstein 1996).
Bracken, M.B., Triche, E.W., Belanger, K., Hellenbrand, K., Leaderer,
B.P., 2003. Association of maternal caffeine consumption with decrements in fetal growth. Am. J. Epidemiol. 157, 456–466. [Discussed also
in 7. Fetal Growth]
This prospective cohort study included 2291 pregnant women
enrolled by 24 gestational weeks (14.4 weeks on average). All
women consuming P150 mg/day during the previous week were
invited to participate as well as a random sample of those consuming <150 mg/day. Caffeine exposures were quantified by measuring caffeine concentrations in urine at the baseline interview,
and by self-reported consumption during the first and last trimester of pregnancy. A major strength of this study is that it was designed to assess pregnancy outcomes in relation to caffeine use
and, thus, incorporated a detailed assessment of caffeinated beverage intake including method of preparation, brand names and
more accurate reporting of serving sizes. The omission of caffeine
from food and medicinal sources, however, may have underestimated exposure status.
Preterm births were not differentiated according to spontaneous or medically indicated deliveries. The frequencies of preterm
birth and IUGR in this cohort were lower than those reported for
the general US population within a similar time period suggesting
that participants consisted of women who had generally healthier
pregnancies compared to the general US population.
No associations between caffeine use and preterm birth were
observed in analyses adjusted for smoking and other potential confounders. The lack of association was consistently observed for urinary caffeine concentrations as well as self-reported use.
Tough, S.C., Newburn-Cook, C.V., White, D.E., Fraser-Lee, M.J.,
Faber, A.J., Frick, C., Svenson, L.W., Sauve, R., 2003. Do maternal characteristics and past pregnancy experiences predict preterm delivery
among women aged 20–34? J. Obstet. Gynaecol. Can. 25, 656–66.
This population-based case-control study of 323 preterm deliveries and 664 controls explored numerous demographic, medical
and lifestyle characteristics as predictors of preterm delivery.
Phone interviews were conducted 4–8 weeks after delivery. Specific details of the methods for caffeine assessment were not provided. Analyses were limited to crude odds ratios assessed
separately for coffee, tea and caffeinated soft drink consumption
measured as <1 cup per day compared to P1 cup per day. A crude
association between coffee consumption and preterm birth was
observed (OR = 1.4, 95% CI 1.0–1.9), but was not maintained in
the final multivariate model reported for statistically significant
predictors of preterm delivery.
consumption during pregnancy in women with type 1 diabetes. J. Matern. Fetal Neonatal Med. 15, 44–50. [Discussed also in 3. Spontaneous Abortion and 6. Congenital Malformations]
In this study of 191 diabetic pregnancies, the monthly assessment of caffeine exposure was limited, measured as the average
number of 8 oz cups consumed daily from all caffeinated beverages, giving equal weighting to coffee, tea and soft drinks. Analyses
of preterm birth focused on caffeine exposure after 20 weeks of
gestation. To assess the quality of this measurement, the authors
compared the serving counts averaged across the last half of pregnancy to estimates of total caffeine consumption during this time
period converted to equivalent cups of coffee per day. The agreement between methods, as measured by the Kappa statistic, was
reported to be 0.61, 0.65 and 0.71 for 0, 1–2, and P3 cups/day.
Thus, the analyses of preterm birth are subject to considerable
exposure misclassification. The authors elected not to report results using total caffeine consumption.
In this report, no association was observed between caffeine
consumption after 20 weeks of gestation and gestational age at
delivery after controlling for age, smoking, glycemic control and
other indicators of diabetes severity. Without a more accurate
measurement of caffeine exposure, the findings offer little insight
into associations with preterm birth.
Santos, I.S., Matijasevich, A., Valle, N.C.J., 2005. Maté drinking during pregnancy and risk of preterm and small for gestational age birth.
J. Nutr. 135, 1120–1123. [Discussed also in 7. Fetal Growth]
2563
This retrospective cohort study explored consumption of maté,
a popular beverage in South America prepared by steeping leaves
of Ilex paraguayensis in hot water. It is considered a major source
of caffeine intake among women in Southern Brazil. Women
(n = 5189) in five local maternity hospitals were interviewed within 24 h following delivery. Maté use during pregnancy was quantified as number of days per week (0, 1–6 and 7 days per week)
rather than number of servings. The amount of actual caffeine consumption was not specifically investigated, but according to results
from a previous study in the same source population (Santos
et al.,1998), women reporting daily consumption of maté averaged
approximately 300 mg of caffeine per day.
Gestational age was measured using a clinical estimate at birth
(i.e., Dubowitz score), which has been shown to have reasonable,
but imperfect, agreement with ultrasound measurements (Vik
et al., 1997). In light of these limitations, the authors’ failure to find
a relationship between maté consumption and duration of pregnancy has limited significance.
Chiaffarino, F., Parazzini, F., Chatenoud, L., Ricci, E., Tozzi, L.,
Chiantera, V., Maffioletti, C., Fedele, L., 2006. Coffee drinking and risk
of preterm birth. Eur. J. Clin. Nutr. 60, 610–613.
This case-control study compared 520 women who delivered at
least three weeks before term to 1966 controls who delivered at
term in the same hospitals in Northern Italy. Because of potential
etiologic heterogeneity, subgroups of preterm births with and
without small for gestational age (SGA) were evaluated separately.
It is unclear how the exposure information was originally collected and combined for analyses. For example, it is not known
whether consumption patterns were reported for specific weeks
or by trimester of pregnancy which were then summed or whether
participants reported an overall estimate of average consumption
during the entire pregnancy. Caffeine consumption during pregnancy was relatively low in this population, limiting the opportunity to assess associations with high intake levels. Given
exposure assessment occurred within the 3 days following delivery, recall bias was a potential concern. Positive associations with
caffeinated beverage intake, however, were not observed.
Associations with tea, cola or decaffeinated coffee were not observed for either preterm subgroup. Women who consumed two or
more servings of coffee per day were at reduced risk of delivering
an SGA newborn before term compared to non-consumers [OR:0.5
(0.3–0.8)]. The reduced risk did not achieve statistical significance
for preterm delivery of an appropriate for gestational age infant
[OR:0.8 (0.6–1.1)].
The authors attribute this unexpected inverse association to potential increases in coffee consumption among controls that may
accompany decreased nausea during the third trimester, in conjunction with decreased coffee consumption among concerned
case mothers who may have become aware of restricted fetal
growth by the third trimester. Pregnancy symptoms, however,
were not evaluated. Given the limitations presented above, this
study does not provide convincing evidence either for or against
a causal link between caffeine and preterm birth.
Bech, B.H., Obel, C., Henriksen, T.B., Olsen, J., 2007. Effect of reducing caffeine intake on birth weight and length of gestation: randomised controlled trial. BMJ 334, 409–414. [Discussed also in 7. Fetal
Growth]
This study distinguishes itself from observational studies by
using an experimental design to randomly assign 1197 heavy coffee drinkers to caffeinated or decaffeinated instant coffee during
the last half of pregnancy. Mean birth weight and gestational age
were compared across treatment groups. While random exposure
assignment helped reduce selection biases, exposure measurement
error remained, since participants in both treatment groups received no restrictions on amount of coffee consumed and were free
to consume other sources of coffee or caffeinated beverages. In
2564
addition, based on interview data, more than one-third of the women in the decaffeinated group were daily consumers of caffeinated coffee (P1 cup/day). This cross-over bias serves to make the
caffeine distribution in the two treatment groups more similar,
thereby weakening the ability to detect true differences should
they exist.
Analysis of the data using the intent to treat approach (i.e., by
treatment assignment) found that caffeinated coffee consumers
had a distribution of pregnancy duration that was similar to that
of decaffeinated coffee consumers. This approach, however,
ignores the individual level exposure data. In the absence of an
analysis of actual caffeine intake, the authors do offer analyses
stratified by ‘‘compliance” which was defined according to frequency of consumption of caffeinated coffee outside of that provided by the study. The lack of association between caffeine
assignment and preterm birth was consistently observed across
compliance groups.
Mikkelsen, T.B., Osterdal, M.L., Knudsen, V.K., Haugen, M., Meltzer,
H.M., Bakketeig, L., Olsen, S.F., 2008. Association between a Mediterranean-type diet and risk of preterm birth among Danish women: a prospective cohort study. Acta Obstet. Gynecol. Scand. 87, 325–330.
Haugen, M., Meltzer, H.M., Brantsaeter, A.L., Mikkelsen, T., Osterdal,
M.L., Alexander, J., Olsen, S.F., Bakketeig, L., 2008. Mediterranean-type
diet and risk of preterm birth among women in the Norwegian Mother
and Child Cohort Study (MoBa): a prospective cohort study. Acta Obstet. Gynecol. Scand. 87, 319–324.
These two companion studies were coordinated to assess the effects of a Mediterranean-type diet on preterm birth in two large
birth cohorts in Denmark and Norway. Low coffee consumption,
defined as 62 cups of coffee per day, was evaluated as part of a
Mediterranean-type diet. Both studies were conducted within
large existing birth cohorts and were restricted to non-smoking
women between ages 21 and 38 with a BMI between 19 and 32,
pregnant with singletons, with normal calorie intake and no history of >3 spontaneous abortions. Preterm deliveries were evaluated in both studies according to early (22–34 weeks) and late
(35–36 weeks) preterm births as well as all preterm births combined (<37 weeks of gestation). Both studies reported results specifically for coffee consumption, but the findings for associations
with preterm birth were inconsistent between the two studies.
The Danish study (Mikkelsen et al., 2008) reported data from
35,530 non-smoking women recruited from the Danish National
Birth Cohort during the 25th week of gestation. Coffee consumption during the previous 4-week period was collected as part of a
food frequency questionnaire. Gestational age at birth was primarily obtained from mother’s reported last menstrual period. Coffee
intake 62 cups per day was associated with a 26% lower odds of
early preterm birth [OR:0.7 (0.6–0.9)] compared to women consuming >2 cups/day, after adjustment for other components of
the Mediterranean diet and other confounders. No association with
late preterm birth was observed [OR:0.9 (0.8, 1.1)].
Because these studies were designed to assess the effects of a
Mediterranean diet and not caffeine intake specifically, measurements of all sources of caffeine exposure were not obtained. Thus,
the results are specific to coffee consumption and not caffeine. To
the extent that (1) coffee is the primary source of caffeine exposure
in Danish women and (2) consumption patterns during the fifth
month of pregnancy represent either the critical window of exposure or patterns of caffeine use maintained during the critical window, the results could be suggestive of a weak effect of caffeine on
early but not late preterm birth.
The study published by Haugen et al. (2008) was conducted
among 26,563 women participating in the Norwegian Mother
and Child Cohort Study. Food frequency questionnaires were
administered during mid-pregnancy (17–24 weeks), but in this
study the reference period was the entire pregnancy up to that
point, rather than the previous month as reported in Mikkelsen
et al. (2008). Thus, coffee exposure was subject to misclassification
that may occur when intake patterns are retrospectively reported
over many months. Information on gestational age at birth was
collected by linking to the Norwegian Medical Birth Registry. In
the Norwegian sample, consuming two or less cups of coffee a
day was not associated with a reduced risk of delivering before
the 35th week [OR:1.11 (0.83, 1.49)] or during the 35th and 36th
weeks [OR:1.15 (0.90, 1.46)].
The inconsistent results of these parallel studies by the same
group of investigators may be attributed to differences in the quality of exposure and outcome measurements or they may be due to
chance. Overall, convincing support for a role of caffeine in the etiology of preterm birth was not demonstrated.
6. Congenital malformations
6.1. Heterogeneity
All malformations are not etiologically identical. Even those
studies limited to one organ or related structures (e.g., lip and palate) are likely evaluating heterogeneous entities. Some studies
have assessed single malformations (Mongraw-Chaffin et al.,
2008; Torfs and Christianson, 2000; Browne et al., 2007; Miller
et al., 2009; Slickers et al., 2008) or utilized more refined sub-classifications of observed phenotypes (Browne et al., 2007; Bille et al.,
2007; Johansen et al., 2009; Schmidt et al., 2009; Collier et al.,
2009) in an attempt to reduce etiologic heterogeneity.
6.2. Biased exposure data
Because congenital malformations are rare, the case-control
(case-referent) design is preferred. This design suffers, however,
from potential recall bias, which could result in overestimating
the contribution of caffeine to the occurrence of congenital malformations. Alternative strategies for control selection, such as the use
of controls with other anomalies, have been proposed to assess the
contribution of recall bias, but these methods do not remove the
bias if it is present (Lieff et al., 1999; Hook, 2000; Schlesselman,
1982).
All studies of caffeine and congenital malformations reviewed
in this report selected controls from non-cases or from a random
sample of all births in the population.
6.3. Relevance of time of exposure to disturbed development
While many studies define the exposure interval broadly as the
first trimester, the relevant window of exposure for most congenital malformations is the period of organogenesis. Exposures after
this period are unlikely to be important.
Since early first trimester exposures coincide with the appearance of pregnancy symptoms for many women, reports of caffeine
consumption averaged across the entire first trimester (or longer
periods) might not reflect actual exposures during the relevant
period of fetal development. Furthermore, the retrospective nature
of exposure assessment limits measurement precision for narrowly
defined periods of gestation. Although the mean timing of symptom onset is reported to occur between 5 and 6 weeks after the last
menstrual period (Gadsby et al., 1993; Lawson et al., 2004), over
13% of pregnant women have been reported to experience symptom onset as early as within the 2 weeks following the estimated
time of conception (i.e., 4 weeks after last menstrual period) (Gadsby et al., 1993). Thus, concern for accurate exposure assessment in
the context of symptom-related changes in caffeine consumption
remains relevant for outcomes such as congenital malformations
that result from very early disruptions to fetal development.
Pre-pregnancy caffeine exposure has been used as a surrogate
for consumption in early stages of gestation before pregnancy
symptoms develop. This approach would result in misclassification
of exposure for women who changed their intake patterns either
because they planned their pregnancy or experienced early pregnancy symptoms. The direction of the bias produced by these measurement errors is difficult to predict. If measurement errors are
equally likely among cases and controls, misclassification would
likely underestimate caffeine-malformation associations when
binary measures of exposure are assessed.
6.4. Outcome ascertainment
Recognition of congenital malformations requires the fetus to
survive until birth, or at least until prenatal diagnosis. Even malformations among fetuses that survive to birth, however, are not always evident at delivery and may go undiagnosed. When
studying malformations that are sometimes fatal, selection bias
can occur if exposure leads to higher rates of pregnancy loss in
malformed fetuses, resulting in lower proportions of exposed cases
(Khoury et al., 1989). Failure to identify malformations resulting in
spontaneous or elective abortions may lead to under-ascertainment of cases. Since most studies of congenital malformations
are conducted in collaboration with birth defect registries, the impact of active vs. passive surveillance systems on the completeness
of reporting must also be considered.
6.5. Review of individual studies of congenital malformations
Natsume, N., Kawai, T., Ogi, N., Yoshida, W., 2000. Maternal risk
factors in cleft lip and palate: case control study. Br. J. Oral Maxillofac.
Surg. 38, 23–25.
This case-control study included 306 cases of cleft lip, cleft palate, or both matched to 306 controls by ‘‘same district during the
same time period”. Given the limited information provided, potential for selection bias cannot be disregarded.
The study was not designed specifically to evaluate caffeine as a
risk factor; however, coffee consumption was part of an assessment of dietary preferences. Details regarding the definition of
the reference periods ‘‘before” and ‘‘during” pregnancy were omitted. The authors do not offer a statistical comparison between
cases and controls beyond assessing the difference in proportions
(chi square test) consuming <1 cup of coffee per week. By our calculations, the crude odds ratios for coffee consumption during
pregnancy were 0.8 (0.6–1.1) for 1–2 cups/week and 0.9 (0.4–
2.3) for 3–6 cups per week, using the <1 cup/week group as the referent. For coffee consumed before pregnancy, the crude odds ratios
were 0.7 (0.5–1.0) for 1–2 cups per week and 0.5 (0.3–0.9) for 3–6
cups per week compared to consumers of <1 cup/week. Thus, case
mothers were less likely to consume coffee before pregnancy than
the unaffected controls. Given the lack of consideration for confounders and great potential for exposure misclassification and recall bias, these data do not make a meaningful contribution to the
body of evidence regarding caffeine and congenital malformations.
consumption during pregnancy in women with type 1 diabetes. J. Matern. Fetal Neonatal Med. 15, 44–50. [Discussed also in 3. Spontaneous Abortion and 5. Gestational Age and Preterm Birth]
This study of pregnancy complications among 191 pregnant
women with type 1 diabetes evaluated associations between first
trimester caffeine use and all major malformations combined.
Any consumption of caffeine during the first trimester (i.e., none
vs. one or more cups of coffee, tea or soft drinks) was not associ-
2565
ated with major malformations [crude OR:2.0 (0.4–11.2)]. The
number of observed malformations was not described, but was
reportedly too small to estimate adjusted odds ratios. This investigation was severely limited by small sample size, potential confounding, the combination of etiologically heterogeneous
malformations, and inadequate exposure assessment.
Torfs, C.P., Christianson, R.E., 2000. Effect of maternal smoking and
coffee consumption on the risk of having a recognized Down syndrome
pregnancy. Am. J. Epidemiol. 152, 1185–1191.
This population-based case-control study included 997 Down
syndrome cases from the California Birth Defects Monitoring Program and 1007 liveborn non-malformed controls from the general
population, frequency matched to cases by hospital of birth. Women were interviewed approximately 5–6 months following delivery for consumption of coffee, tea and soft drinks ‘‘around the time
of conception”. The primary analyses focused on coffee consumption, with the reference group being women who consumed 0–3
cups of coffee per day.
A protective association between heavy coffee intake (P4 cups/
day compared to 63 cups/day) and Down syndrome was observed
among non-smokers [OR:0.5 (0.3–0.8)], but not smokers [OR:1.6
(0.8–3.4)]. The authors interpret their results as evidence that
non-smoking mothers (defined as not smoking within three
months of conception) consuming high levels of caffeine were
more likely to miscarry a fetus with Down syndrome, thereby
reducing the prevalence of cases recognized at later deliveries
among heavy caffeine users. In other words, the authors suggest
that the observed protective effect may indicate selection bias
inherent in studies of congenital malformations such as Down syndrome which go largely undetected due to early spontaneous abortions (as described by Khoury et al., 1989). The authors speculate
that the lack of a similar observed effect among smokers could
be attributed to increased metabolism and caffeine clearance associated with smoking.
While the results presented in this report offer no evidence for a
role of caffeine in the etiology of Down syndrome, they also do not
directly evaluate contributions of caffeine use to the early loss of
Down syndrome fetuses.
Bille, C., Olsen, J., Vach, W., Knudsen, V.K., Olsen, S.F., Rasmussen, K.,
Murray, J.C., Andersen, A.M., Christensen, K., 2007. Oral clefts and life
style factors – a case-cohort study based on prospective Danish data.
Eur. J. Epidemiol. 22, 173–181.
This study utilized prospectively collected data on coffee, tea
and cola consumption in a case-cohort design, which included
134 cases of cleft lip with and without cleft palate and 58 cases
with cleft palate only identified within the Danish National Birth
Cohort. Controls (n = 828) were randomly selected from the birth
cohort. The authors observed no associations with coffee intake
for all oral clefts combined or by subtype. Mothers of babies with
isolated cleft palate had 2.5 (1.1–5.6) times greater odds of
consuming 5 or more cups of tea per day compared to mothers
of controls. Weekly cola intake exceeding one liter was marginally
associated with cleft lip with or without cleft palate [OR:1.5 (0.9–
2.4)].
The major strength of this study is the evaluation of oral clefts
by clinically confirmed subtypes (i.e., cleft lip with and without
cleft palate and cleft palate only), which may be etiologically
distinct.
There are several limitations, however. Analyses by subtype
were limited by small numbers and lacked precision. While exposure assessment was reported to take place between gestational
weeks 12–27 (for 90%), few details of the data collection are presented. It is not clear whether consumption was assessed as usual
consumption since conception or as average consumption during
the first trimester or during a specific reference period. There
was no attempt to assess caffeine intake from all sources combined
2566
or patterns of caffeine consumption that may fluctuate with pregnancy symptoms. Closure of lip and palate structures occurs
around the 8th week of gestation during normal fetal development
(Burdi and Faist, 1967; Yoon et al., 2000). Thus the relevant period
of exposure for oral clefts coincides with the time in early pregnancy when pregnancy symptoms begin to appear, with average
onset between 5 and 6 weeks of gestation and symptoms peaking
by week 8 on average (Gadsby et al., 1993; Lawson et al., 2004).
Thus, reports of average daily consumption across the entire first
trimester may not reflect actual patterns of exposure during the
etiologically relevant period of fetal development.
The absence of consistent results by type of beverage is not consistent with an etiologic role for caffeine in the development of
cleft lip and/or palate. Although the measurement errors described
above may have tended to obscure associations with caffeinated
beverages, these errors would not be expected to differ by type
of beverage. While the non-statistically significant protective effect
of coffee may be interpreted as evidence for caffeine-related demise among malformed fetuses, we caution against this conclusion
since oral clefts do not commonly result in early fetal loss and thus
are not susceptible to the form of selection bias described by Khoury et al. (1989) that occurs when cases are identified from live
births.
Browne, M.L., Bell, E.M., Druschel, C.M., Gensburg, L.J., Mitchell,
A.A., Lin, A.E., Romitti, P.A., Correa, A., 2007. National Birth Defects Prevention Study. Maternal caffeine consumption and risk of cardiovascular malformations. Birth Defects Res. A Clin. Mol. Teratol. 79, 533–543.
This large case-control study from the National Birth Defects
Prevention Study identified specific cardiovascular malformations
(n = 4196) from the birth defect registries of eight states and
3957 controls randomly selected from hospital records or birth
certificates within each state. This population-based study has a
number of strengths including a sample size large enough to evaluate specific malformations considered etiologically homogeneous
and an assessment of caffeine intake from multiple sources. The
authors also evaluated a complete and well justified list of potential confounders and effect modifiers, providing specific details
regarding the criteria for assessment. After conducting a thorough
analysis, the authors reported no positive associations between
pre-pregnancy caffeine intake and cardiovascular malformation
subtypes.
The primary weakness of this study involves retrospective
exposure assessment far removed in time from the critical window
of susceptibility. Interviews were conducted 8–12 months after
delivery on average, with some occurring as late as 24 months after
delivery dates. While the burden of recall appeared to be similar
for cases and controls, the excessive time gap could have led to
reporting errors. Although sensitivity analyses conducted by the
authors showed no difference in results when subjects interviewed
more than 1 year after the estimated date of delivery were eliminated, these findings do not alleviate concerns about exposure misclassification since recall up to 1 year after delivery could be just as
flawed.
The rationale provided for evaluating pre-pregnancy exposure
rather than first trimester exposure is that it may better depict intake patterns during the susceptible period for fetal heart development, which tends to occur before most pregnancy symptoms
develop (Lacroix et al., 2000). The authors acknowledge that exposure misclassification would likely occur for pregnancy planners or
those experiencing early pregnancy symptoms who changed their
intake patterns during this early period of gestation.
In the authors’ words, the study ‘‘does not provide any appreciable evidence of an association between maternal caffeine consumption and risk of [cardiovascular malformations]”.
Mongraw-Chaffin, M.L., Cohn, B.A., Cohen, R.D., Christianson, R.E.,
2008. Maternal smoking, alcohol consumption, and caffeine consump-
tion during pregnancy in relation to a son’s risk of persistent cryptorchidism: a prospective study in the Child Health and Development
Studies cohort, 1959–1967. Am. J. Epidemiol. 167, 257–261.
This nested case-control study of cryptorchidism was conducted among children born to mothers enrolled in the Child
Health and Development Studies between 1959 and 1967. Cases
(n = 84) were boys with an undescended testicle at birth that remained undescended until 2 years of age. Controls (n = 252) were
selected from non-cases and matched 3:1 by race/ethnicity and
date of birth. The authors report a modest association between
cryptorchidism and caffeine use equivalent to 3 cups of coffee
per day [OR:1.4 (1.1–1.9)] after controlling for alcohol, smoking,
body mass index and child’s birth weight.
The strengths of the study include prospective data collection
on health behaviors at a time when there was less stigma associated with the use of tobacco, alcohol and caffeine during pregnancy. Thus, under-reporting of such patterns would be less
likely to occur in this study population and exposure reports would
not be influenced by knowledge of the birth outcome.
Selection bias cannot be ruled out as an alternative explanation
for the weak association observed in this study. A large proportion
of cases (33/101 = 33%) and controls (40/252 = 16%) were excluded
from analyses because of missing data. If exclusions differed from
inclusions with respect to case/control status and factors that correlated with caffeine consumption, the observed association with
caffeine could be distorted.
Another issue pertaining to study quality is the potential for
exposure misclassification. Caffeine exposure was assessed by
interview conducted ‘‘during early pregnancy”, but the specific reference period was not described and the assessment did not capture fluctuations that typically occur during pregnancy. Thus, it is
questionable whether the recorded data reflects accurate exposures during the relevant period of fetal development. Although
the susceptible period for development of cryptorchidism is not
well understood, the relevant time window could plausibly include
mid to late pregnancy, given normal testicular descent occurs between gestational weeks 25 and 32 (Rotondi et al., 2001).
Potential confounding by gestational diabetes was also not considered. Mild gestational diabetes has been identified previously as
a risk factor of cryptorchidism (Virtanen et al., 2006) while coffee
consumption during pregnancy has been linked to a reduced risk
of gestational diabetes (Adeney et al., 2007). Thus, gestational diabetes may serve as a negative confounder of the association between caffeine and cryptorchidism. Thus, in the presence of a
true association, adjusting for the influence of gestational diabetes
may potentially strengthen the observed association.
Given the limitations considered, the study results do not provide convincing evidence for an association between early pregnancy caffeine use and persistent cryptorchidism.
Slickers, J.E., Olshan, A.F., Siega-Riz, A.M., Honein, M.A., Aylsworth,
A.S., for the National Birth Defects Prevention Study, 2008. Maternal
body mass index and lifestyle exposures and the risk of bilateral renal
agenesis or hypoplasia: The National Birth Defects Prevention Study.
Am. J. Epidemiol. 168, 1259–1267.
Also originating from the National Birth Defects Prevention
Study (NBDPS), this multicenter case-control study examined caffeine intake and other maternal characteristics as risk factors for
bilateral renal agenesis or renal hypoplasia. The study consisted
of 75 cases and 868 controls. Utilizing data on typical daily caffeine
intake reported for the year preceding pregnancy, the study shares
many of the main strengths and limitations of Browne et al. (2007)
and other caffeine studies derived from the NBDPS. However, the
authors incorporated additional data on reported changes in coffee,
tea or soda consumption during pregnancy (more, same, less or no
intake compared to the year before pregnancy) in an effort to more
accurately classify ‘‘negligible” vs. ‘‘nonnegligible” intake during
pregnancy. Intake was classified as negligible when there was no
report of coffee, tea or soda consumption during pregnancy, mean
daily caffeine intake was <10 mg in the year before pregnancy and
no intake was reported during pregnancy, or mean daily caffeine
intake was <50 mg in the year before pregnancy accompanied by
decreased consumption of one or more caffeine sources and no increases in the other caffeine sources during pregnancy. In this report, contributions from chocolate or caffeine-containing
medications were not included. No associations with nonnegligible
caffeine intake were observed (adjusted OR:1.01 (95% CI 0.58–
1.75). The authors expressed concern that exposure misclassification due to reliance on caffeine exposure recalled for the year before pregnancy may have obscured modest associations.
Miller, E.A., Manning, S.E., Rasmussen, S.A., Reefhuis, J., Honein,
M.A., and the National Birth Defects Prevention Study, 2009. Maternal
exposure to tobacco smoke, alcohol and caffeine, and risk of anorectal
atresia: National Birth Defects Prevention Study 1997–2003. Paediat.
Perinat. Epidemiol. 23, 9–17.
Using data from the National Birth Defects Prevention Study,
this case-control study analyzed 464 infants with anorectal atresia
and 4940 controls with no major birth defects. Mothers were interviewed by telephone 6–24 months after the estimated date of
delivery (mean 228 days). Caffeine consumption was measured
as total caffeine derived from usual intake of beverages and chocolate reported for the year before pregnancy.
Modest borderline significant associations were observed for all
categories of caffeine intake [OR:1.4 (1.0–1.9) for 10–99 mg;
OR:1.3 (1.0–1.8) for 100–299 mg; OR:1.5 (1.0–2.2) for P300 mg
compared to <10 mg]. It is notable that the point estimates were
strengthened (but less stable) when caffeine exposure was refined
according to reported changes in caffeine intake during pregnancy
(i.e., restricted to those in reference group reporting same or less
caffeine during pregnancy and those in the exposed groups reporting same consumption or more during pregnancy) [OR:2.6 (1.2–
5.6) for 10–99 mg; OR:2.1 (1.0–4.4) for 100–299 mg; OR:2.6
(1.2–6.0) for P300 mg]. All findings are reported as unadjusted because none of the factors evaluated met the criterion for confounding (i.e., 10% change in caffeine estimate), including smoking.
However, ‘‘any smoking” during the periconceptional period was
found to be associated with anorectal atresia in these data, which
raises the question of residual confounding by smoking due to
measurement error or variable specification in the multivariate
model. Recall bias and exposure misclassification resulting from
the time gap for reporting or from poor representation of true periconceptional caffeine exposures are other considerations for the
observed associations.
Johansen, A.M.W., Wilcox, A.J., Lie, R.T., Andersen, L.F., Drevon, C.A.,
2009. Maternal consumption of coffee and caffeine-containing beverages and oral clefts: a population-based case-control study in Norway.
Am. J. Epidemiol. 169, 1216–1222.
Cases of cleft lip with or without cleft palate (n = 377) and cleft
palate only (n = 196) were identified from two surgical centers in
Norway. Controls (n = 763) were randomly selected from the national birth registry. Mothers were interviewed 14–15 weeks after
delivery for coffee, tea and soft drink consumption during the first
three months of pregnancy. No associations were observed for total caffeine intake from all sources and risk of cleft lip with or
without cleft palate [OR:1.2 (0.7–2.0) for P500 mg compared to
0–100 mg] or cleft palate only [OR:1.1 (0.5–2.2) for P500 mg compared to 0–100 mg]. Coffee intake, however, was weakly associated
with cleft lip with or without cleft palate [OR:1.6 (1.1–2.4) for P3
cups/day], but not cleft palate only. In contrast, tea consumption
appeared to protect against risk of both subtypes.
The study controlled for known or suspected confounders,
including smoking and nausea. The primary limitation, however,
was the possibility of recall bias. As noted for Bille et al. (2007),
2567
the lack of association with total caffeine intake and inconsistent
results by type of beverage is not supportive of an etiologic role
for caffeine in the development of cleft lip and/or palate.
Schmidt, R.J., Romitti, P.A., Burns, T.L., Browne, M.L., Druschel, C.M.,
Olney, R.S., and the National Birth Defects Prevention Study, 2009.
Maternal caffeine consumption and risk of neural tube defects. Birth
Defects Res. A Clin. Mol. Teratol. 85, 879–889.
Due to the large size of the National Birth Defects Prevention
Study, this case-control study was able to separately assess associations with specific types of neural tube defects including spina
bifida (n = 459), anencephaly (n = 218) and encephalocele (n = 91)
as compared to 4143 non-malformed controls. Because total average daily caffeine intake from coffee, tea, soda and chocolate was
assessed during the year prior to pregnancy, the use of caffeinecontaining medications during the periconceptional period (one
month before pregnancy through the first three months of pregnancy) was assessed separately. Caffeine exposure was evaluated
as total mg/day and by source as cups per day. Modest associations
with spina bifida were observed for any consumption of caffeine
(P10 mg/day; OR = 1.4, 95% CI 1.1–1.9), any caffeinated coffee
(P1 cup/month; OR = 1.3; 95% CI 1.0–1.6), and any caffeinated
soda (>0/day; OR = 1.2; 95% CI 1.0–1.6). When stratified by smoking, alcohol and maternal age, the associations between spina bifida and any caffeine intake were only observed among women
without the high risk characteristics (i.e., among non-smokers,
non-alcohol users, younger aged women). Any consumption of caffeinated tea was found to be protective for spina bifida (OR = 0.7;
95% CI 0.6–0.9). When examined across increasing categories of
consumption, associations were limited to the groups with lower
levels of consumption; thus, no evidence of a dose–response with
increasing caffeine intake was observed. Similarly, associations
with encephalocele were observed for coffee and tea, but only for
categories of 1 cup/day of coffee and not greater levels of consumption. No associations with anencephaly were observed.
As with other NBDPS studies, the authors acknowledge the potential for recall bias (e.g., interviews conducted an average of 9.7
and 8.0 months after delivery for cases and controls), measurement error (e.g., exposure reported for year before pregnancy,
and used only the implied serving size of ‘‘a cup”) and possible
residual confounding (e.g., imprecise measurement of some covariates such as smoking (yes/no)). Although the lack of a dose–response detracts from the strength of evidence for a causal
association, the authors suggest tolerance effects that develop
among regular users may diminish dose effects.
Collier, S.A., Browne, M.L., Rasmussen, S.A., Honein, M.A., and the
National Birth Defects Prevention Study, 2009. Maternal caffeine intake during pregnancy and orofacial clefts. Birth Defects Res. A Clin.
Mol. Teratol. 85, 842–849.
The National Birth Defects Prevention Study was also the source
for this case-control study of orofacial clefts. This study included
1531 infants with cleft lip with or without cleft palate, 813 infants
with cleft palate only and 5711 controls with no major birth defects. Isolated and multiple malformations were also assessed separately. The NBDPS details concerning caffeine measurement and
other study considerations have been described above. For total
caffeine intake from coffee, tea, soda and chocolate, odds ratios
were modestly elevated for limited amounts of caffeine intake
(10–199 mg/day) relative to <10 mg/day for most outcomes (e.g.,
isolated cleft lip with or without cleft palate: OR = 1.2; 95% CI
1.0–1.5 for 100–<200 mg/day), but no associations were observed
for higher levels of caffeine intake (200 <300 mg/day and
300 + mg/day). When intake was examined by source, three or
more cups of coffee per day was protective against cleft palate only
with multiple unrelated malformations (OR = 0.3; 95% CI 0.1–0.9)
but an increased odds ratio was observed for similar tea intake
(OR = 2.4; 95% CI 1.3–4.6). Medication containing 100 + mg of caf-
2568
feine per dose was associated with isolated cleft lip with or without cleft palate (OR = 2.3; 95% CI 1.3–4.0) relative to no use of caffeine-containing medication. The authors interpret their findings
as failing to provide support for an overall association between
maternal caffeine intake and orofacial clefts, noting that effects
with caffeine-containing medications should be further explored
for the role played by other substances, confounding by indication
and the possibility that users represent those with higher daily caffeine intake.
7. Fetal growth restriction
As a perinatal outcome of interest, fetal growth is considered a
marker of healthy intrauterine development and a predictor of
postnatal morbidity and mortality (Savitz et al., 2002). Most studies of caffeine and fetal growth have assessed intrauterine growth
restriction (IUGR) (also referred to as SGA) defined as birth weight
<10th centile for gestational age according to a standard growth
curve from a selected reference population appropriate for the infant (by sex and race). Other measures of fetal growth explored in
relation to caffeine use include birth weight, relative birth weight
(Z-scores), low birth weight (<2500 g), high birth weight
(>4000 g), birth length, ponderal index (birth weight (g)/birth
length (cm)3 100), head circumference, abdominal circumference, placental weight, and placental diameter.
7.1. Defining growth restriction
The external standards of birth weight for gestational age used
to define IUGR vary among studies. While it is agreed that a single
standard is not appropriate for use in all populations and that the
standard used should be population-specific, no population-specific standards have been adopted (Ott, 2006). Furthermore, the
birth weight standards in use do not stratify the growth curves
by the same population characteristics, so they may be inconsistently specified by any combination of sex, race and/or parity.
Because it is difficult to distinguish constitutionally small babies from babies who are genuinely growth restricted, births defined as IUGR using a population standard (such as the 10th
centile) will capture heterogeneous outcomes. Thus, the inclusion
of genetically small but otherwise normal babies as IUGR would
likely attenuate associations of potential risk factors with fetal
growth restriction. Alternative methods have been proposed for
identifying infants who fail to obtain their inherent growth potential. These methods compare birth weight to a customized fetal
growth curve, which predicts weight for gestational age according
to maternal height and weight, sex, race/ethnicity and parity
(Gardosi, 1997, 2006; Zhang et al., 2007). Customized growth
curves have not been widely used, however, in studies of IUGR
etiology.
7.2. Accurate estimation of gestational age
The issues addressed in Section 5.2 concerning accurate estimation of gestational age and preterm birth are also relevant to studies of IUGR. While ultrasound-based pregnancy dating is often
preferred, this method is also derived from measurements of fetal
growth (e.g., the bi-parietal diameter of the skull), which can be
influenced by fetal growth restriction. Thus, IUGR infants may escape identification if this circular argument causes gestational
age to be underestimated by the use of ultrasound measurements.
7.3. Relevant window of susceptibility
The timing of exposure assessment is important for studying
the etiology of fetal growth restriction, but the exact window of
susceptibility is unknown. Although fetal size was previously believed to be determined during the third trimester of pregnancy
when weight gain is most rapid, recent evidence suggests that fetal
growth restriction may be determined by conditions occurring
early during the first trimester of pregnancy (Smith et al., 1998;
Smith, 2004; Bukowski et al., 2007). This uncertainty underscores
the importance of assessing caffeine exposure throughout the
course of pregnancy when investigating possible associations with
fetal growth.
Although the role of the pregnancy signal is well acknowledged
in studies of caffeine and spontaneous abortion, confounding by
pregnancy symptoms has been considered less often in studies of
IUGR. Yet, the pregnancy signal might be related to placental size.
Fetal growth restriction is often associated with a small placenta
(Naeye, 1987; Redline and Patterson, 1994; Thame et al., 2004).
Still unknown is whether the process that limits fetal growth also
limits placental growth, or whether a small placenta reduces fetal
growth by synthesizing inadequate amounts of proteins that promote growth. Either way, the small placenta can be expected to
synthesize less of the hormones needed for growth, some of which
at high levels may produce the pregnancy signal (Lawson et al.,
2002).
7.5. Review of individual studies of fetal growth restriction
Grosso, L.M., Rosenberg, K.D., Belanger, K., Saftlas, A.F., Leaderer, B.,
Bracken, M.B., 2001. Maternal caffeine intake and intrauterine growth
retardation. Epidemiology 12, 447–455.
This analysis evaluated IUGR among 2714 women delivering
singleton, live births within the Yale Health in Pregnancy Study
(1988–1991). The investigators observed no association between
IUGR and caffeine intake during the first or seventh month of pregnancy. Stratification by smoking status did not change the results.
Gestational age was estimated by administering the Ballard
examination within the first day of life (Ballard et al., 1979), which
may have contributed to nondifferential misclassification of IUGR.
Although the study population was large, the study was limited by
the low percentage of women consuming more than moderate
amounts of caffeine (>300 mg/day) during the two time periods assessed (5% in month 1 and 2% in month 7). This limited the power
to detect associations with caffeine use >300 mg/day. On the other
hand, the authors acknowledge that recall bias was a possibility for
month seven caffeine consumption, which was reported after
delivery. In addition, pregnancy symptoms were not considered
as a potential confounder.
The data presented in this study provide no evidence to support
a link between caffeine use in the first or seventh month of pregnancy and fetal growth restriction.
Clausson, B., Granath, F., Ekbom, A., Lundgren, S., Nordmark, A.,
Signorello, L.B., Cnattingius, S., 2002. Effect of caffeine exposure during
pregnancy on birth weight and gestational age. Am. J. Epidemiol. 155,
429–436. [Discussed also in 5. Gestational Age and Preterm Birth]
In this study, 873 controls from a case-control study of spontaneous abortion (Cnattingius et al., 2000) were followed to investigate the effects of first and third trimester caffeine consumption on
pregnancy outcomes, including birth weight, Z-scores and gestational age. Weekly caffeine intake through the first 6 weeks of
pregnancy was obtained from the first trimester interview (conducted between weeks 6–12), while biweekly intake from 7 weeks
of gestation onward was obtained from the third trimester interview (conducted between weeks 32–34). The long recall period
associated with the retrospective reports of consumption occurring
up to 7 months prior may have produced nondifferential misclassification. Furthermore, averaging intake across large spans of time
such as trimesters or throughout pregnancy may misrepresent
peak exposures during the relevant window of susceptibility.
The authors reported no differences in adjusted mean birth
weight or Z-scores across caffeine intake categories defined as
0–99, 100–299, 300–499 or P500 mg/day during the first, second
or third trimesters or for average consumption throughout pregnancy. The authors controlled for confounding by third trimester
smoking (yes/no using cotinine >15 ng/ml) and pregnancy symptoms, but this had little impact on the results. Furthermore, no
effect modification by smoking or pregnancy symptoms was
observed.
Klebanoff, M.A., Levine, R.J., Clemens, J.D., Wilkins, D.G., 2002.
Maternal serum caffeine metabolites and small-for-gestational age
birth. Am. J. Epidemiol. 155, 32–37. [Discussed also in 5. Gestational
Age and Preterm Birth]
This study evaluated third trimester serum paraxanthine concentrations in archived samples from Collaborative Perinatal Project participants. Caffeine metabolites might measure biologic
dose better than reported caffeine consumption, but reflect only
recent exposures (Aldridge et al., 1981). Because the serum samples were stored for more than 30 years before being analyzed,
the stability of the caffeine metabolite in the available specimens
is also questioned.
The authors reported that risk of delivering an SGA infant increased with rising third trimester serum paraxanthine concentrations, but only among smokers. The increased risk among smokers
was modest (displayed graphically as ORs 2.0 and lower) and
only present for categories of paraxanthine concentrations exceeding the 65th percentile (>715 ng/ml). These effects were observed
after controlling for self-reported number of cigarettes smoked
per day within the stratum of smokers. No associations with serum
caffeine concentrations were observed. The authors acknowledge
that residual confounding due to under-reporting of number of cigarettes smoked could have exaggerated observed associations for
paraxanthine. Furthermore, pregnancy symptoms were not evaluated. Because the modest associations observed in this study could
be influenced in an unpredictable manner by sample desiccation
and confounding, cautious interpretation is advised.
Balat, O., Balat, A., Ugur, M.G., Pençe, S., 2003. The effect of smoking
and caffeine on the fetus and placenta in pregnancy. Clin. Exp. Obstet.
Gynecol. 30, 57–59.
To evaluate the effect of caffeine intake on newborn and placental characteristics, the authors recruited a group of women who
smoked <10 cigarettes per day (n = 60) and a group of non-smokers
(n = 63) who delivered at term (37–41 weeks). The reference period for the reported caffeine intake from coffee and tea was not
specified. The authors reported lower mean birth weights and placental weights for women reporting caffeine intake of >300 mg/
day compared to women consuming <300 mg/day. Mean differences by caffeine intake were greater for smokers (187 g difference
in birth weight and 97 g difference in placental weight) than nonsmokers (128 g difference in birth weight and 81 g difference in
placental weight). No differences in mean length, head circumference or placental diameter were observed in either group.
The mean differences reported in this study were based on
small numbers (e.g., n = 17 smokers consuming >300 mg/day;
n = 19 non-smokers consuming >300 mg/day), resulting in unstable point estimates. Other than stratification by smoking status,
this study did not attempt to control for important confounders
such as maternal age, alcohol use, gestational age at birth or
amount smoked per day. Exposure misclassification is also likely
2569
given use of a single measurement of caffeine intake, which did
not capture sources other than tea and coffee or specify cup size.
Thus, the results do not offer convincing support for an effect of
caffeine on fetal growth or placental development.
Bracken, M.B., Triche, E.W., Belanger, K., Hellenbrand, K., Leaderer,
B.P., 2003. Association of maternal caffeine consumption with decrements in fetal growth. Am. J. Epidemiol. 157, 456–466. (Discussed also
in 5. Gestational Age and Preterm Birth)
This prospective cohort study of 2291 pregnant women 624
gestational weeks evaluated urinary caffeine concentrations and
self-reported caffeine exposures during early and late pregnancy
in relation to IUGR, low birth weight and preterm birth. A major
strength of this study is that it was designed to assess pregnancy
outcomes in relation to caffeine consumption and, thus, incorporated a very detailed assessment of caffeine exposure from beverage sources.
First and third trimester caffeine consumption was not associated with increased risk of IUGR or low birth weight. Increased
concentrations of urinary caffeine (mg/g creatinine) were also
not associated with IUGR [OR:1.0 (0.8–1.1)] but appeared to be
protective for low birth weight [OR:0.7 (0.5–1.0)]. When birth
weight and caffeine were assessed as continuous variables, each
100 mg of caffeine consumed during the first trimester was associated with a birth weight reduction of 28 g (95% CI 10–46 g). No
such associations were observed with low birth weight [OR:1.00
(0.98–1.00)]. The authors conclude that moderate caffeine consumption during the first trimester may have a modest effect on
birth weight that may not be clinically important. No associations
with third trimester caffeine consumption were reported. Ultimately, the authors’ reasoned that urinary caffeine may not be a
useful biomarker since less than 2.0% of caffeine is excreted in this
form.
This study provides no evidence to suggest that caffeine consumption during early or late pregnancy is related to growth
restriction as measured by IUGR and low birth weight. Only modest reductions in birth weight were observed with increasing caffeine use.
Ørskou, J., Henriksen, T.B., Kesmodel, U., Secher, N.J., 2003. Maternal characteristics and lifestyle factors and the risk of delivering high
birth weight infants. Obstet. Gynecol. 102, 115–120.
This study of risk factors for high birth weight (>4000 g) evaluated a large cohort of women without diabetes who gave birth in
Denmark (1990–1999).
Women seeking
prenatal care
(n = 24,093) completed a questionnaire at approximately 16 weeks
of gestation reporting average daily consumption of caffeinated
beverages. The reference period for the self-reported caffeine intake was not described.
Consuming more than 200 mg/day was associated with a decreased odds of giving birth to a high birth weight infant compared
to consuming <200 mg/day [OR:0.9 (0.8–1.0) for 200–399 mg/day
and OR:0.9 (0.8–1.0) for P400 mg/day.
The main strengths of this study were the large sample size and
control for number of cigarettes smoked per day (e.g., non-smoker,
1–4, 5–9, 10–14, 15 + cigarettes/day), alcohol intake, gestational
age, and other confounders. Misclassification of caffeine, however,
was possible given the single assessment at 16 weeks of gestation
would not capture changing patterns of consumption during pregnancy. The modest protective odds ratio may also overestimate the
relative risk given the outcome was relatively common (19%) in
this study population.
Vik, T., Bakketeig, L.S., Trygg, K.U., Lund-Larsen, K., Jacobsen, G.,
2003. High caffeine consumption in the third trimester of pregnancy:
gender-specific effects on fetal growth. Paediatr. Perinat. Epidemiol.
17, 324–331.
This study population included subjects (n = 858) from the Norwegian site of a multi-site Scandinavian study of SGA. This study is
2570
one of few to prospectively collect caffeine exposure data through
the use of food diaries. Diaries were completed across 3 weekdays
in the second (17–20 weeks) and third trimesters (33 weeks).
While likely to capture current intake during the two specified
time points with fewer recall errors, food diaries would not necessarily reveal intake patterns that may fluctuate from week to week,
month to month, or even weekday to weekend. For analyses, caffeine was dichotomized to represent intake above and below the
mean value (232 mg/day at 17 weeks; 205 mg/day at 33 weeks).
No association was observed between ‘‘high” caffeine intake at
17 weeks and birth of an SGA infant [OR:1.1 (0.6–2.1)], but high
consumption at 33 weeks was associated with an increased odds
of SGA [OR:1.6 (1.0–2.5)]. The association for third trimester caffeine consumption was modified by infant sex [OR:2.8 (1.4–5.5)
for males; OR:1.0 (0.5–2.0) for females]. These estimates were adjusted for smoking at conception (yes/no), pregnancy weight, education and previous SGA birth, but not for maternal age. Although
the findings were consistent among smokers and non-smokers,
residual confounding by amount smoked could remain if reported
caffeine consumption reflected actual frequency of smoking
(among smokers and inaccurately reported non-smokers alike)
(Morrison, 1984).
The sex-specific effects of caffeine on SGA are difficult to explain. Fetal sex might be a marker of more severe pregnancy symptoms. For example, having a female fetus has been strongly
associated with hyperemesis gravadarum (Tan et al., 2006). We
do not know of any sex differences for less severe nausea, but if
similar patterns exist, women with male fetuses may have had less
nausea, and therefore consumed more coffee. The possibility of
sex-specific effects of caffeine on SGA is intriguing, but consideration of all potential confounders including nausea and aversions
would be necessary before a causal link could be legitimately
considered.
Parazzini, F., Chiaffarino, F., Chatenoud, L., Tozzi, L., Cipriani, S.,
Chiantera, V., Fedele, L., 2005. Maternal coffee drinking in pregnancy
and risk of small for gestational age birth. Eur. J. Clin. Nutr. 59, 299–
301.
This case-control study of term singletons born in Italy selected
555 SGA babies and 1966 controls. The methods for estimating
gestational age for the SGA definition were not described. Information on coffee, tea, and cola was collected after delivery, measured
as number of cups per day before pregnancy and during each trimester. Total caffeine intake from all beverage sources was not
assessed.
The authors observed no associations between SGA and intake
of three or more cups of coffee during the first, second or third trimester of pregnancy. Likewise, no associations were observed for
heavy coffee consumption (P4 cups/day) before pregnancy
[OR:1.3 (0.9–1.9)]. Because the control inclusion criterion for full
term deliveries was not applied equally to the selection of cases,
the authors repeated their analyses after restricting cases to births
delivered after 37 weeks of gestation. This reduced the odds ratios
for the heaviest coffee consumption categories to exactly 1.0 for all
time periods assessed (i.e., before pregnancy and during each trimester). When tea, cola and decaffeinated coffee were each evaluated separately, no associations were observed.
Santos, I.S., Matijasevich, A., Valle, N.C.J., 2005. Maté drinking during pregnancy and risk of preterm and small for gestational age birth.
J. Nutr. 135, 1120–1123. [Discussed also in 5. Gestational Age and
Preterm Birth]
In addition to evaluating preterm birth, this retrospective cohort study assessed SGA births in relation to exposure to a popular
caffeinated beverage consumed in South America called maté. All
women (n = 5189) giving birth to singletons in the five local maternity hospitals were interviewed within the 24 h following delivery.
Frequency of maté use during pregnancy was assessed as number
of days the drink was consumed per week (0, 1–6 and 7 days per
week). The authors equated daily maté consumption with a daily
average caffeine intake of 300 mg (Santos et al., 1998). Gestational
age was estimated using the Dubowitz score.
After controlling for maternal age, education, parity, cigarette
smoking during the third trimester, previous abortion and previous
low birth weight infant, the prevalence of SGA births did not differ
among women consuming maté 0, 1–6 or 7 days per week. The primary limitation of this study was the non-specific measure of caffeinated beverage consumption in days per week rather than
servings per day. Other sources of caffeine intake were not considered. Recall bias was also possible given women were interviewed
after delivery. Confounding by alcohol and nausea was not considered in the analyses. Because of its limitations, this study should be
viewed as weak evidence that caffeine does not influence the risk
of fetal growth restriction.
Grosso, L.M., Triche, E.W., Belanger, K., Benowitz, N.L., Holford, T.R.,
Bracken, M.B., 2006. Caffeine metabolites in umbilical cord blood,
cytochrome P-450 1A2 activity, and intrauterine growth restriction.
Am. J. Epidemiol. 163, 1035–1041.
This study was conducted within the prospective cohort described by Bracken et al. (2003) who were selected on the basis of
caffeine consumption Por <150 mg/day. Of the 2291 participants,
1606 had cord blood samples available for analysis of serum caffeine,
paraxanthine, theophylline and theobromine concentrations.
No associations were observed for the highest quartile of serum
caffeine concentrations [OR:0.5 (0.1–2.0)]. Caffeine metabolites
were not associated with IUGR when evaluated independently,
but infants born to women with the highest concentrations of
paraxanthine (fourth quartile) had a 3-fold increased risk of IUGR
[OR:3.3 (1.2–9.2)], but only when also controlling for serum caffeine concentrations. This led the authors to suspect that metabolic
activity rather than absolute concentrations of caffeine and its
metabolites may have more relevance for fetal growth, so the
remaining analyses assessed the ratio of paraxanthine to caffeine
as a marker of CYP1A2 enzymatic activity. The adjusted odds ratio
for a one standard deviation increase in the ratio of paraxanthine to
caffeine was 1.2 (1.1–1.4). Thus, fast metabolizers experienced a
higher risk of IUGR than slow metabolizers.
The major strength of this study was the use of biomarkers in
cord blood measurements as indicators of biologic dose entering
the fetal circulation. Given the short half-life of caffeine and its
metabolites, concentrations observed at delivery may not accurately characterize exposure throughout pregnancy unless caffeine
consumption habits were fairly consistent over time.
Only 55% of the population had gestational age confirmed by
ultrasound. As with all studies of IUGR, errors in gestational age
based on inaccurate recall of the date of the last menstrual period
can lead to misclassification of IUGR. Residual confounding by
smoking could have overestimated the association between caffeine metabolism and IUGR. Urinary cotinine levels (before
25 weeks of gestation) were available for the cohort and reported
in a previous paper by Bracken et al. (2003), but only self-reported
smoking (yes/no) during the third trimester was utilized in these
analyses. This choice likely reflects the preference for third trimester smoking measurements more proximate to the timing of the
cord blood measurements.
As the authors acknowledge, CYPIA2 activity may be a marker
for other compounds such as tobacco smoke that induce metabolic
activity and alter fetal growth (Campbell et al., 1987b). The observed effect of metabolic activity, however, was similar in self-reported smokers and non-smokers. Confounding by nausea was not
evaluated, but it is unknown if pregnancy symptoms would be related to metabolic activity.
These findings are suggestive of a possible modest relationship
between high metabolic activity, as measured by the ratio of para-
xanthine to caffeine concentrations in cord serum, and risk of
IUGR. Better control for smoking would have enhanced confidence
in the results.
Tsubouchi, H., Shimoya, K., Hayashi, S., Toda, M., Morimoto, K.,
Murata, Y., 2006. Effect of coffee intake on blood flow and maternal
stress during the third trimester of pregnancy. Int. J. Gynaecol. Obstet.
92, 19–22.
This study did not directly assess the effects of caffeine on fetal
growth, but measured the effects of a one-cup dose of coffee on
maternal and fetal blood flow during the third trimester. A group
of 10 non-smoking women in their third trimester of pregnancy
underwent Doppler examinations before and after coffee consumption (100 mg caffeine). All participants typically consumed
1–3 cups of coffee per day. Resistance index values were calculated
to measure blood flow in the maternal uterine artery, umbilical artery and fetal middle cerebral artery 30 min after coffee intake.
Coffee had no effect on maternal or fetal blood flow. When the effects of coffee on stress levels were assessed by measuring salivary
cortisol and chromogranin A (CgA), coffee intake reduced salivary
cortisol levels in the 10 pregnant women but not in a comparison
group of 14 women who were not pregnant. CgA concentrations,
which were considered an indication of physiological stress, increased following coffee consumption in non-pregnant women,
but significant increases were not detectable in pregnant women.
The small study size and limited statistical power restricts confidence in the absence of associations.
Bech, B.H., Obel, C., Henriksen, T.B., Olsen, J., 2007. Effect of reducing caffeine intake on birth weight and length of gestation: randomised controlled trial. BMJ 334, 409–414. [Discussed also in 5.
Gestational Age and Preterm Birth]
Women who regularly drank at least three cups of coffee per
day (n = 1197) were randomly assigned to caffeinated or decaffeinated instant coffee during the last half of pregnancy. The pregnancies were followed to evaluate differences in gestational age and
mean birth weight. Although participants were provided unlimited
amounts of coffee, either caffeinated or decaffeinated as assigned,
they were also free to consume other sources of coffee and caffeinated beverages. According to self-reports, more than one-third of
those assigned to decaffeinated coffee consumed P1 cup of caffeinated coffee on a daily basis.
Using the intent to treat approach to analysis, no difference in
birth weight was observed when comparing women assigned to
caffeinated vs. decaffeinated coffee. This approach, however,
would misclassify true exposure status given poor compliance
with assigned coffee treatment. When analyses were restricted
to subgroups defined as women reporting no consumption of
other caffeinated coffee (n = 283) or women reporting <1 cup of
other caffeinated coffee per day (n = 266), the results were unchanged. Birth weight remained similar across caffeine assignment groups within non-smokers and within women smoking
1–10 cigarettes per day. However, among women smoking P10
cigarettes per day, those randomized to caffeinated coffee had a
lower mean birth weight than the decaffeinated group (mean
difference: 263 g (97–430 g), adjusted for parity, pre-pregnancy
BMI and length of gestation. The possible interaction between
smoking and caffeine consumption is unconvincing without presentation of the results by compliance or, preferably, by actual
caffeine consumption.
Diego, M., Field, T., Hernandez-Reif, M., Vera, Y., Gil, K., GonzalezGarcia, A., 2007. Caffeine use affects pregnancy outcome. J. Child Adolesc. Subst. Abuse 17, 41–49.
In this study, 750 pregnant women between 20 and 28 weeks
gestation were recruited from a prenatal clinic to study depression,
anxiety and substance use. Birth outcomes including birth weight
and ‘‘pregnancy complications” were collected from medical charts
on a sub-sample of 452 participants. Pregnancy complications
2571
were not defined. Caffeine consumption was collected from interviews conducted at enrollment and described as caffeinated drinks
per day during pregnancy, but the data collection procedures were
not described. It would appear, however, that all caffeinated beverages were weighted equally. The range of exposure was reported as
0–6 caffeinated drinks per day, with 38% reporting no caffeine consumption. Weak negative correlations were observed between
number of caffeinated drinks and birth weight (r = 0.11,
p < 0.05). In hierarchical linear regression analyses restricted to
women who reported no smoking or alcohol use, caffeine intake
explained 8% of the variability in birth weight when controlling
for anxiety and depression. In light of the study limitations – which
include questionable exposure assessment and lack of control for
other important confounders such as maternal age and body mass
– this study provides limited information.
Infante-Rivard, C., 2007. Caffeine intake and small-for-gestationalage birth: modifying effects of xenobiotic-metabolising genes and
smoking. Paediatr. Perinat. Epidemiol. 21, 300–309.
This hospital-based, case-control study of SGA births identified
451 cases matched to an equal number of controls by gestational
age, sex and race. Interviews were generally conducted within
two days of delivery. CYP1A2 and CYP2E1 polymorphisms were
genotyped using peripheral blood samples from mothers and newborns. Genotypes were dichotomized as those with one or two copies of the variant allele vs. those with the wild type.
The authors observed no association between caffeine consumption and SGA when using either continuous or dichotomous
(<300 vs. P300 mg/day) measures of caffeine intake during the
month before pregnancy or during the first, second, or third trimester of pregnancy. Adjustment for smoking (none vs. any) and
nausea had little impact on the observed lack of associations,
although stratification by smoking status revealed a 22% increased
odds of SGA among non-smokers for every 100 mg/day consumed
during the first trimester [OR:1.2 (1.0–1.5)]. The effect of first trimester caffeine consumption on SGA was not modified by maternal or newborn genetic polymorphisms. Birth weight was
reduced by 31 (95% CI 61, 1) and 38 g (95% CI 68, 8) for
every 100 mg of caffeine consumed during the second and third
trimester, respectively. The effects of caffeine on birth weight were
also restricted to non-smokers.
Although the study improved upon most previous studies by
including measurements of nausea by trimester (presence vs. absence), residual confounding would persist if severity or aversions
were the influential factors. The study does not provide evidence of
effect modification of caffeine use on SGA by CYP1A2 or CYP2E1
polymorphisms in mother or child. Because the modest effects
on SGA and birth weight among non-smokers could potentially
be due to recall bias or inaccurately reported smoking status, the
strength of the evidence for an effect of caffeine on fetal growth
restriction is limited.
Xue, F., Willett, W.C., Rosner, B.A., Forman, M.R., Michels, K.B., 2008.
Parental characteristics as predictors of birthweight. Hum. Reprod. 23,
168–177.
This cross-sectional study was conducted among the Nurses’
Mother’s Cohort. Mothers of the Nurses’ Health Study participants
(n = 34,063) were sent questionnaires to collect information on
daughter’s birth weight as well as behaviors and conditions during
the pregnancy. Mothers were 60–80 years old when completing
the questionnaire about events that occurred approximately 40–
60 years earlier. Thus, the long recall period is the major limitation
of this study. Coffee consumption was measured as intake ‘‘during
pregnancy” in cups/day. Daughter’s birth weight was also self-reported, although a subset were validated against birth certificates
(r = 0.85).
Birth weight was negatively associated with coffee consumption during pregnancy, decreasing by 15, 34 and 54 g for consump-
2572
tion of 1–2, 3–4 and P5 cups of coffee per day. The odds of IUGR
were modestly increased in a dose–response fashion with increasing coffee consumption, with the strongest association observed
among women reporting P5 cups of coffee per day [OR:1.6 (1.3–
2.1)]. Although many important confounders were considered,
pregnancy symptoms were not accounted for. The weak associations observed in this study could potentially be attributed to measurement error and confounding.
CARE Study Group, 2008. Maternal caffeine intake during pregnancy and risk of fetal growth restriction: a large prospective observational study. BMJ 337, a2332.
This prospective study from the United Kingdom evaluated a
cohort of 2635 women recruited early in pregnancy (8–12 weeks
of gestation). A major strength of the study was its thorough
assessment of caffeine exposure. Three questionnaires were
administered to capture recall of caffeine consumption for each trimester (5–12, 13–28, and 29–40 weeks of pregnancy). The questionnaire considered all beverage, food, and over-the-counter
medication sources, brands, portion sizes, and methods of preparation. Caffeine half-life was measured in saliva samples as an indicator of fast or slow caffeine clearance, collected one and five
hours following a 63.5 mg caffeine challenge (500 ml diet soda
over 20 min) after fasting.
This study also strived to improve upon IUGR classification by
applying customized fetal growth curves (i.e., identifying birth
weight <10th percentile for gestational age according to maternal
height, weight, ethnicity, parity and sex) in an effort to avoid misclassifying constitutionally small infants as growth restricted. Furthermore, gestational age was confirmed by ultrasound in all
pregnancies. Another positive feature was the use of salivary cotinine concentrations to identify current smokers, non-smokers and
those exposed to second-hand smoke. However, these measurements were taken during the first trimester recruitment visit and
may not accurately reflect tobacco exposure throughout pregnancy. Caffeine consumption averaged over the entire pregnancy
was modestly associated with IUGR [OR:1.2 (0.9–1.6) for 100–
199 mg/day; OR:1.5 (1.1–2.1) for 200–299 mg/day; OR:1.4 (1.0–
2.0) for P300 mg/day compared to 6100 mg/day]. Similar associations were observed for caffeine exposure in each trimester, with
slightly larger odds ratios for the 2nd and 3rd trimester. Associations between consumption >200 mg/day and reduced birth
weight in the range of 60–70 g were also observed across all time
periods. When the data were stratified by caffeine clearance, the
association with IUGR appeared to remain only among the fast
metabolizers; although, the confidence intervals in each strata largely overlapped with the exception of the 200–299 mg/day category (test for interaction p = 0.06).
The major limitation of this study is the potential for confounding by pregnancy symptoms and aversions. The authors
mention an assessment of the effects of adjusting for nausea,
but these results were not presented and no description of the
nature of the nausea data was provided. A related concern is
the extremely low participation rate (20%). The authors dismiss
the likelihood of selection bias, but the higher prevalence of IUGR
in the study population (13%) compared to the general population (10%) suggests that women with a history of fetal growth
restriction may have been more motivated to participate in the
study. If pregnancies destined to be growth restricted produced
a weaker pregnancy signal, selection into the study would be related to both the outcome (IUGR) and the exposure (higher caffeine intake due to fewer symptoms and aversions). Thus, the
more highly exposed, growth restricted group would be overrepresented. Because pregnancy symptoms and aversions were
not controlled in these analyses, selection bias remains a plausible explanation for the modest associations observed in this
study.
The authors of the study considered that, for the first time
among similar studies, the quantification of caffeine from all
known sources reflected ‘‘a true picture of total caffeine intake
by women during pregnancy”. However, when the validation study
compared the caffeine questionnaire to a 3-day food dairy and repeated salivary caffeine and paraxanthine concentrations, only fair
to moderate levels of agreement (intraclass correlation for food
diary = 0.5; Kappa for biomarkers 0.33–0.65) were observed (Boylan et al., 2008). Although considerable steps toward improving
the assessment of caffeine exposure were taken, the data collection
instrument does not succeed in eliminating all concerns about caffeine measurement error. Potential for recall bias specific to 2nd
and 3rd trimester exposure reports may also exist, as women became aware of restricted fetal growth during routine mid- or late
pregnancy ultrasounds. This could explain the slightly stronger
associations observed for caffeine consumption during 13–28 and
29–40 weeks of gestation.
8. Discussion/conclusions
8.1. Subfecundity
Of the nine publications since 2002, one evaluated multiple outcomes associated with fertility treatment, one considered self-reported ovulatory infertility, three addressed time to conception,
and most (4) assessed the relationship between caffeine and semen
parameters.
The only study to assess the effect of caffeine on endpoints of
assisted reproductive technology reported no influence of previous
or current caffeine intake on oocyte retrieval, fertilization, embryo
transfer or the occurrence of a clinical pregnancy. The effect of limited caffeine use on failure to achieve a live birth was inconsistent
for exposures reported for different time periods that reflected
usual and recent exposures. Caffeine use around the week of IVF
or GIFT procedures was not associated with failure to achieve a live
birth, while associations with usual lifetime use and use reported
during first clinic visit were observed. Replication of results in larger populations undergoing ART is needed to address concerns
about statistical power, precision, residual confounding and caffeine exposures >50 mg/day.
Exposure measurement errors are a primary concern for the few
recent studies addressing time to conception and ovulatory infertility. Potential recall bias and exposure misclassification may explain the modest association reported for coffee and tea
consumption and increased time to pregnancy. No support for an
association with infertility due to ovulation disorders was provided, but exposure measurement error was likely introduced as
a result of the timing of exposure assessments.
Evaluations of semen quality have consistently failed to observe
adverse effects associated with caffeine intake. Studies of DNA
damage, however, have been more limited but inconsistent. Most
of the studies of male reproductive outcomes have suffered from
lack of detailed reporting of caffeine exposure assessment, potential exposure misclassification for the relevant etiologic window,
no or limited control for confounders, potential selection bias, or
restriction to fertile men, which limited the ability to detect caffeine-related abnormalities.
In summary, consistent relationships between caffeine intake
and measures of subfecundity have not been observed.
8.2. Spontaneous abortion
The current evidence remains insufficient to permit conclusions
regarding the potential role of caffeine in spontaneous abortion.
Studies of caffeine and spontaneous abortion are complicated by
the challenge of separating cause and effect.
Studies have not successfully addressed the complex interrelationship between viability, pregnancy signal symptoms and caffeine consumption patterns. Of the 15 studies that have
evaluated caffeine and spontaneous abortion since the Leviton
and Cowan (2002) review, 10 did not attempt to control for pregnancy signal symptoms. Although positive associations were consistently reported by these 10 studies, these consistent findings
may very well be attributed to bias that persists across all studies.
As a marker of pregnancy viability and a correlate of exposure, confounding by pregnancy signal symptoms may explain observed
associations with caffeine use. Women with pregnancies that go
to term experience more frequent and severe nausea early in pregnancy compared to women whose pregnancies end in spontaneous
abortion (Weigel and Weigel, 1989). Lawson et al. (2002, 2004)
demonstrated that weekly duration of nausea (in hours) and appetite loss (in days) is positively related to human chorionic gonadotropin (hCG) levels during pregnancy, whereas hCG levels are also
negatively related to coffee consumption. Of those women who decrease coffee consumption during the first trimester of pregnancy,
65% report a physical aversion to coffee (Lawson et al., 2004). Thus,
women experiencing viable pregnancies tend to experience a
stronger pregnancy signal that ultimately drives caffeine intake
downward. The pregnancy signal is, therefore, a crucial confounder
of the association between caffeine and pregnancy viability.
Results from studies that have attempted to control for nausea
and vomiting during pregnancy have been less consistent. Improved assessments of relevant symptom characteristics, such as
aversions to taste and smell, in addition to symptom severity,
duration, frequency and timing are needed to improve study validity. The study by Wen et al. (2001) provides perhaps the best evidence for the pregnancy signal phenomenon to date, whereby
increased risk of spontaneous abortion was only observed for caffeine consumed after nausea onset, but not for caffeine consumed
before nausea onset or among those without nausea. Other persistent problems with the validity of studies of caffeine and spontaneous abortion include confounding by smoking and potential recall
bias.
8.3. Fetal death
Three of the four studies evaluating caffeine and fetal death reported moderately positive associations of similar magnitude.
Three of the four were conducted by members of the same research
group using similar methodologies and similar study populations.
None, however, sufficiently address concerns regarding confounding by pregnancy symptoms. Only one study attempted to control
for pregnancy symptoms, but the limited assessment of the pregnancy signal defined as presence or absence of nausea or vomiting
during the first trimester was insufficient to avoid residual confounding. However, Bech et al. (2005) provided a useful demonstration of how observed associations can be inflated by
undetected fetal demise. As with studies of spontaneous abortion,
the interpretation of this body of work, which has consistently reported modest associations across studies, needs to consider that
these studies may also share common sources of bias which may
explain the observed relationship with caffeine use.
8.4. Preterm birth
Larger studies considering total caffeine exposure consistently
reported no increased risk of delivery before 37 weeks of gestation.
No studies since 2000, however, distinguished spontaneous from
medically indicated preterm deliveries. Two studies of Mediterranean-type diets (Mikkelsen et al., 2008; Haugen et al., 2008) sepa-
2573
rately evaluated early and late preterm births, with inconsistent
results between studies. Coffee intake limited to 62 cups/day
was linked to a lower odds of preterm delivery 634 weeks of gestation, but not late preterm delivery (35–36 weeks) in the Danish
National Birth Cohort (Mikkelsen et al., 2008). A similar study conducted within the Norwegian Mother and Child Cohort observed
no relationship between coffee intake and early or later preterm
delivery (Haugen et al., 2008). Combining etiologically distinct outcomes could obscure associations within clinical subtypes of preterm birth.
8.5. Congenital malformations
With a few exceptions, recent studies have not reported an increased risk of malformations with greater caffeine consumption.
However, the body of evidence for any single malformation or subgroup is limited. The reports of modest associations between coffee
intake and cryptorchidism and total caffeine intake and anorectal
atresia could not confidently rule out potential sources of bias such
as selection bias due to missing data, exposure misclassification
and confounding.
8.6. Fetal growth
Studies of caffeine and fetal growth restriction are equivocal,
with approximately half of the studies in this review reporting
weak associations with intrauterine growth restriction or reduced
birth weight and half observing no effects. The strength of the evidence for a potential effect of caffeine on fetal growth restriction is
diminished by the inability to rule out alternative, credible explanations for the observed associations, namely confounding by
pregnancy symptoms and aversions.
8.7. Summary
In conclusion, the weight of evidence does not support a positive relationship between caffeine consumption and adverse reproductive or perinatal outcomes. On the whole, associations with
subfecundity, preterm delivery and congenital malformations are
not routinely observed. Studies of pregnancy loss and fetal growth
have generated more interest due to the frequency with which
adverse effects are reported in connection with caffeine use. Our
review identifies significant methodological weaknesses common
to studies of spontaneous abortion, fetal death and fetal growth
restriction, which limit confidence in causal interpretation. Consistent with the conclusion of the previous review (Leviton and
Cowan, 2002), the studies available from January 2000 through
December 2009 do not provide convincing evidence that caffeine
consumption increases risk of any reproductive adversity. Future
studies addressing the methodological limitations of current
research may alter this conclusion. In particular, quantitative
methods are available for adjustment for measurement error in
the absence of a gold standard (Joseph et al. 1995) and would be
especially useful for estimating the impact of errors in the assessment of caffeine exposure.
Conflict of Interest
The authors declare that there are no conflicts of interest.
Acknowledgements
This work was supported by the Caffeine Working Group of the
North American Branch of the International Life Sciences Institute
(ILSI). ILSI North America is a public, non-profit foundation that
2574
provides a forum to advance understanding of scientific issues related to the nutritional quality and safety of the food supply by
sponsoring research programs, educational seminars and workshops, and publications. ILSI North America receives support primarily from its industry membership. Dr. Peck received a grant
from the organization for her work reviewing, analyzing, and summarizing the information contained in this article. ILSI North
America also received funding for this project from the National
Coffee Association. The opinions expressed herein are those of
the authors and do not necessarily represent the views of either
funding organization.
References
Adeney, K.L., Wiliams, M.A., Schiff, M.A., Chunfang, Q., Sorensen, T.K., 2007. Coffee
consumption and risk of gestational diabetes. Acta Obstet. Gynecol. Scand. 86,
161–166.
Aldridge, A., Bailey, J., Neims, A.H., 1981. The disposition of caffeine during and after
pregnancy. Semin. Perinatol. 5, 310–314.
Arnaud, M.J., 1994. Metabolism of caffeine and other components of coffee. In:
Garattini, S. (Ed.), Caffeine, Coffee and Health. Raven Press, New York, pp. 43–
95.
Baird, D.D., Wilcox, A.J., Weinberg, C.R., 1986. Use of time to pregnancy to study
environmental exposures. Am. J. Epidemiol. 124, 470–480.
Balat, O., Balat, A., Ugur, M.G., Pençe, S., 2003. The effect of smoking and caffeine on
the fetus and placenta in pregnancy. Clin. Exp. Obstet. Gynecol. 30, 57–59.
Ballard, J.L., Novak, K.K., Driver, M.A., 1979. A simplified score for assessment of fetal
maturation of newly born infants. J. Pediatr. 95, 769–774.
Ballard, J.L., Khoury, J.C., Wedig, K., Wang, L., Eilers-Walsman, B.L., Lipp, R., 1991.
New Ballard Score, expanded to include extremely premature infants. J. Pediatr.
119, 417–423.
Barone, J.J., Roberts, H.R., 1996. Caffeine consumption. Food Chem. Toxicol. 34, 119–
129.
Bech, B.H., Nohr, E.A., Vaeth, M., Henriksen, T.B., Olsen, J., 2005. Coffee and fetal
death, a cohort study with prospective data. Am. J. Epidemiol. 162, 983–990.
Bech, B.H., Autrup, H., Nohr, E.A., Henriksen, T.B., Olsen, J., 2006. Stillbirth and slow
metabolizers of caffeine, comparison by genotypes. Int. J. Epidemiol. 35, 948–
953.
Bech, B.H., Obel, C., Henriksen, T.B., Olsen, J., 2007. Effect of reducing caffeine intake
on birth weight and length of gestation, randomised controlled trial. BMJ 334,
409–414.
Berhman, R.E., Butler, A.S. (Eds.), 2007. Preterm birth. Causes, Consequences, and
Prevention. The National Academies Press, Washington, DC.
Bille, C., Olsen, J., Vach, W., Knudsen, V.K., Olsen, S.F., Rasmussen, K., Murray, J.C.,
Andersen, A.M., Christensen, K., 2007. Oral clefts and life style factors – a casecohort study based on prospective Danish data. Eur. J. Epidemiol. 22, 173–181.
Boylan, S.M., Cade, J.E., Kirk, S.F.L., Greenwood, D.C., White, K.L.M., Shires, S.,
Simpson, N.A.B., Wild, C.P., Hay, A.W.M., 2008. Assessing caffeine exposure in
pregnant women. Br. J. Nutr. 100, 875–872.
Bracken, M.D., Triche, E., Grosso, L., Hellenbrand, K., Belanger, K., Leaderer, B.P.,
2002. Heterogeneity in assessing self-reports of caffeine exposure, implications
for studies of health effects. Epidemiology 13, 165–171.
Bracken, M.B., Triche, E.W., Belanger, K., Hellenbrand, K., Leaderer, B.P., 2003.
Association of maternal caffeine consumption with decrements in fetal growth.
Am. J. Epidemiol. 157, 456–466.
Browne, M.L., Bell, E.M., Druschel, C.M., Gensburg, L.J., Mitchell, A.A., Lin, A.E.,
Romitti, P.A., Correa, A.National Birth Defects Prevention Study, 2007. Maternal
caffeine consumption and risk of cardiovascular malformations. Birth Defects
Res. A Clin. Mol. Teratol. 79, 533–543.
Bukowski, R., Smith, G.C., Malone, F.D., Ball, R.H., Nyberg, D.A., Comstock, C.H.,
Hankins, G.D., Berkowitz, R.L., Gross, S.J., Dugoff, L., Craigo, S.D., Timor-Tritsch,
I.E., Carr, S.R., Wolfe, H.M., D’Alton, M.E. FASTER Research Consortium, 2007.
Fetal growth in early pregnancy and risk of delivering low birth weight infant,
prospective cohort study. BMJ 334, 836–840.
Burdi, A.R., Faist, K., 1967. Morphogenesis of the palate in normal human embryos
with special emphasis on the mechanisms involved. Am. J. Anat. 120, 149–160.
Campbell, M.E., Grant, D.M., Inaba, T., Kalow, W., 1987a. Biotransformation of
caffeine, paraxanthine, theophylline, and theobromine by polycyclic aromatic
hydrocarbon-inducible cytochrome(s) in human liver microsomes. Drug Metab.
Dispos. 15, 237–249.
Campbell, M.E., Spielberg, S.P., Kalow, W., 1987b. A urinary metabolite ratio that
reflects systemic caffeine clearance. Clin. Pharmacol. Ther. 42, 157–165.
CARE Study Group, 2008. Maternal caffeine intake during pregnancy and risk of
fetal growth restriction, a large prospective observational study. BMJ 337,
a2332.
Carrillo, J., Benitez, J., 2000. Clinically significant pharmacokinetic interactions
between dietary caffeine and medications. Clin. Pharmacokinet. 39, 127–153.
Chavarro, J.E., Rich-Edwards, J.W., Rosner, B.A., Willett, W.C., 2009. Caffeinated and
alcoholic beverage intake in relation to ovulatory disorder infertility.
Epidemiology 20, 374–381.
Chiaffarino, F., Parazzini, F., Chatenoud, L., Ricci, E., Tozzi, L., Chiantera, V.,
Maffioletti, C., Fedele, L., 2006. Coffee drinking and risk of preterm birth. Eur.
J. Clin. Nutr. 60, 610–613.
Clausson, B., Granath, F., Ekbom, A., Lundgren, S., Nordmark, A., Signorello, L.B.,
Cnattingius, S., 2002. Effect of caffeine exposure during pregnancy on birth
weight and gestational age. Am. J. Epidemiol. 155, 429–436.
Cnattingius, S., 2004. The epidemiology of smoking during pregnancy, smoking
prevalence, maternal characteristics, and pregnancy outcomes. Nicotine Tob.
Res. 6 (Suppl. 2), S125–S140.
Cnattingius, S., Signorello, L.B., Anneren, G., Clausson, B., Ekbom, A., Ljunger, F., Blot,
W.J., McLaughlin, J.K., Peterson, G., Rane, A., Granath, F., 2000. Caffeine intake
and the risk of first-trimester spontaneous abortion. N. Engl. J. Med. 343, 1839–
1845.
Cohn, B.A., Overstreet, J.W., Fogel, R.J., Brazil, C.K., Baird, D.D., Cirillo, P.M., 2002.
Epidemiologic studies of human semen quality, considerations for study design.
Am. J. Epidemiol. 55, 664–671.
Cole, D.C., Wainman, B., Sanin, L.H., Weber, J.P., Muggah, H., Ibrahim, S., 2006.
Environmental contaminant levels and fecundability among non-smoking
couples. Reprod. Toxicol. 22, 13–19.
Collier, S.A., Browne, M.L., Rasmussen, S.A., Honein, M.A.the National Birth Defects
Prevention Study, 2009. Maternal caffeine intake during pregnancy and
orofacial clefts. Birth Defects Res. A Clin. Mol. Teratol. 85, 842–849.
Dales, L.G., Ury, H.K., 1978. An improper use of statistical significance testing in
studying covariables. Int. J. Epidemiol. 7, 373–375.
Diego, M., Field, T., Hernandez-Reif, M., Vera, Y., Gil, K., Gonzalez-Garcia, A., 2007.
Caffeine use affects pregnancy outcome. J. Child Adolesc. Subst. Abuse 17, 41–
49.
Dubowitz, L.M., Dubowitz, V., Goldberg, C., 1970. Clinical assessment of gestational
age in the newborn infant. J. Pediatr. 77, 1–10.
Ford, R.P., Tappin, D.M., Schluter, P.J., Wild, C.J., 1997. Smoking during pregnancy,
how reliable are maternal self reports in New Zealand? J. Epidemiol.
Community Health 51, 246–251.
Gadsby, R., Barnie-Adshead, A.M., Jagger, C., 1993. A prospective study of nausea
and vomiting during pregnancy. Br. J. Gen. Pract. 43, 245–248.
Gardosi, J., 1997. Customized growth curves. Clin. Obstet. Gynecol. 40, 715–722.
Gardosi, J., 2006. New definition of small for gestational age based on fetal growth
potential. Horm. Res. 65 (Suppl. 3), 15–18.
George, L., Granath, F., Johansson, A.L.V., Olander, B., Cnattingius, S., 2006. Risks of
repeated miscarriage. Paediat. Perinat. Epidemiol. 20, 119–126.
Giannelli, M., Doyle, P., Roman, E., Pelerin, M., Hermon, C., 2003. The effect of
caffeine consumption and nausea on the risk of miscarriage. Paediatr. Perinat.
Epidemiol. 7, 316–323.
Greenland, S., Gustafson, P., 2006. Accounting for independent nondifferential
misclassification does not increase certainty that an observed association is in
the correct direction. Am. J. Epidemiol. 164, 63–68.
Grosso, L.M., Bracken, M.B., 2005. Caffeine metabolism, genetics, and perinatal
outcomes, a review of exposure assessment consideration during pregnancy.
Ann. Epidemiol. 15, 460–466.
Grosso, L.M., Rosenberg, K.D., Belanger, K., Saftlas, A.F., Leaderer, B., Bracken, M.B.,
2001. Maternal caffeine intake and intrauterine growth retardation.
Grosso, L., Triche, E., Bracken, M.B., 2004. Regarding ‘‘An unexpected distribution of
sodium concentration in serum specimens stored for more than 30 years”. Ann.
Grosso, L.M., Triche, E.W., Belanger, K., Benowitz, N.L., Holford, T.R., Bracken, M.B.,
2006. Caffeine metabolites in umbilical cord blood, cytochrome P-450 1A2
activity, and intrauterine growth restriction. Am. J. Epidemiol. 163, 1035–1041.
Hassan, M.A., Killick, S.R., 2004. Negative lifestyle is associated with a significant
reduction in fecundity. Fertil. Steril. 81, 384–392.
Haugen, M., Meltzer, H.M., Brantsaeter, A.L., Mikkelsen, T., Osterdal, M.L., Alexander,
J., Olsen, S.F., Bakketeig, L., 2008. Mediterranean-type diet and risk of preterm
birth among women in the Norwegian Mother and Child Cohort Study (MoBa), a
prospective cohort study. Acta Obstet. Gynecol. Scand. 87, 319–324.
Hook, E.B., 2000. What kind of controls to use in case control studies of malformed
infants, recall bias versus ‘‘teratogen specificity” bias. Teratology 61, 325–326.
Horak, S., Polanska, J., Widlak, P., 2003. Bulky DNA adducts in human sperm,
relationship with fertility, semen quality, smoking, and environmental factors.
Mutat. Res. 537, 53–65.
Infante-Rivard, C., 2007. Caffeine intake and small-for-gestational-age birth,
modifying effects of xenobiotic-metabolising genes and smoking. Paediatr.
Jensen, T.K., Bonde, D.P., Joffe, M., 2006. The influence of occupational exposure on
male reproductive function. Occup. Med. 56, 544–553.
Joffe, M., Villard, L., Li, A., Plowman, R., Vessey, M., 1993. Long-term recall of timeto-pregnancy. Fertil. Steril. 60, 99–104.
Joffe, M., Key, J., Best, N., Keiding, N., Scheike, T., Jensen, T.K., 2005. Studying time to
pregnancy by use of a retrospective design. Am. J. Epidemiol. 162, 115–124.
Johansen, A.M.W., Wilcox, A.J., Lie, R.T., Andersen, L.F., Drevon, C.A., 2009. Maternal
consumption of coffee and caffeine-containing beverages and oral clefts, a
population-based case-control study in Norway. Am. J. Epidemiol. 169, 1216–
1222.
Joseph, L., Gyorkos, T.W., Coupal, L., 1995. Bayesian estimation of disease prevalence
and the parameters of diagnostic tests in the absence of a gold standard. Am. J.
Epidemiol. 141, 263–272.
Kalow, W., Tange, B.-K., 1991. Caffeine as a metabolic probe, exploration of the
enzyme-inducing effect of cigarette smoking. Clin. Pharmacol. Ther. 49, 44–48.
Karypidis, A.H., Söderström, T., Nordmark, A., Granath, F., Cnattingius, S., Rane, A.,
2006. Association of cytochrome P450 1B1 polymorphism with first-trimester
miscarriage. Fertil. Steril. 86, 1498–1503.
Khoury, M.J., Flanders, W.D., James, L.M., Erickson, J.D., 1989. Human teratogens,
prenatal mortality, and selection bias. Am. J. Epidemiol. 130, 361–370.
Khoury, J.C., Miodovnik, M., Buncher, C.R., Kalkwarf, H., McElvy, S., Khoury, P.R.,
Sibai, B., 2004. Consequences of smoking and caffeine consumption during
pregnancy in women with type 1 diabetes. J. Matern. Fetal Neonatal Med. 15,
44–50.
Killick, S., Trussell, J., Cleland, K., Moreau, C., 2009. Factors associated with
subfertility among women attending an antenatal clinic in Hull. Hum. Fertil.
12, 191–197.
Klebanoff, M.A., 1998a. Conceptualizing categories of preterm birth. Prenat.
Neonatal Med. 3, 13–15.
Klebanoff, M.A., Longnecker, M.P., 2004. Response to Letter by Grosso et al. Ann.
Klebanoff, M.A., Shiono, P.H., 1995. Top down, bottom up and inside out, reflections
on preterm birth. Paediatr. Perinat. Epidemiol. 9, 125–129.
Klebanoff, M.A., Levine, R.J., Clemens, J.D., DerSimonian, R., Wilkins, D.G., 1998b.
Serum cotinine concentration and self-reported smoking during pregnancy. Am.
J. Epidemiol. 148, 259–262.
Klebanoff, M.A., Levine, R.J., Dersimonian, R., Clemens, J.D., Wilkins, D.G., 1998c.
Serum caffeine and paraxanthine as markers for reported caffeine intake in
pregnancy. Ann. Epidemiol. 8, 107–111.
Klebanoff, M.A., Levine, R.L., Clemens, J.D., et al., 1999. Maternal serum
paraxanthine, a caffeine metabolite, and the risk of spontaneous abortion. N.
Engl. J. Med. 341, 1639–1644.
Klebanoff, M.A., Levine, R.J., Clemens, J.D., Wilkins, D.G., 2002. Maternal serum
caffeine metabolites and small-for-gestational age birth. Am. J. Epidemiol. 155,
32–37.
Klonoff-Cohen, H., Bleha, J., Lam-Kruglick, P., 2002. A prospective study of the
effects of female and male caffeine consumption on the reproductive
endpoints of IVF and gamete intra-fallopian transfer. Hum. Reprod. 17,
1746–1754.
Lacroix, R., Eason, E., Melzack, R., 2000. Nausea and vomiting during pregnancy: A
prospective study of its frequency, intensity, and patterns of change. Am. J.
Obstet. Gynecol. 182, 931–937.
Lampe, J., King, I., Li, S., Grate, M., Barale, K., Chen, C., Feng, Z., Potter, J.D., 2000.
Brassica vegetables increase and apiaceous vegetables decrease cytochrome
P450 1A2 activity in humans, changes in caffeine metabolic ratios in response to
controlled vegetable diets. Carcinogenesis 21, 1157–1162.
Lawson, C.C., LeMasters, G.K., 2006. Regarding ‘‘Caffeine metabolism, genetics and
perinatal outcomes: a review of exposure assessment considerations during
pregnancy”. Ann. Epidemiol. 16, 733.
Lawson, C.C., LeMasters, G.K., Levin, L.S., Liu, J.H., 2002. Pregnancy hormone
metabolite patterns, pregnancy symptoms, and coffee consumption. Am. J.
Epidemiol. 156, 428–437.
Lawson, C.C., LeMasters, G.K., Wilson, K.A., 2004. Changes in caffeine consumption
as a signal of pregnancy. Reprod. Toxicol. 18, 625–633.
Leviton, A., 1987. Single-cause attribution. Dev. Med. Child Neurol. 29, 805–807.
Leviton, A., Cowan, L., 2002. A review of the literature relating caffeine consumption
by women to their risk of reproductive hazards. Food Chem. Toxicol. 40, 1271–
1310.
Li, D.-K., Weng, X.P., Odouli, R., 2008. Reply to caffeine and miscarriage: case closed?
Am. J. Obstet. Gynecol. 199, e13–e14.
Lieff, S., Olshan, A.F., Werler, M., Savitz, D.A., Mitchell, A.A., 1999. Selection bias and
the use of controls with malformations in case-control studies of birth defects.
Lindqvist, R., Lendahls, L., Tollbom, O., Aberg, H., Håkansson, A., 2002. Smoking
during pregnancy, a comparison of self-reports and cotinine levels in 496
women. Acta Obstet. Gynecol. Scand. 81, 240–244.
Louik, C., Hernandez-Diaz, S., Werler, M.M., Mitchell, A.A., 2006. Nausea and
vomiting in pregnancy, maternal characteristics and risk factors. Paediatr.
Lynch, C.D., Zhang, J., 2007. The research implications of the selection of a
gestational age estimation method. Paediatr. Perinat. Epidemiol. 21 (Suppl. 2),
86–96.
Maconochie, N., Doyle, P., Prior, S., Simmons, R., 2007. Risk factors for first trimester
miscarriage – results from a UK-population-based case-control study. Br. J.
Obstet. Gynaecol. 114, 170–186.
Matijasevich, A., Barros, F.C., Santos, I.S., Yemini, A., 2006. Maternal caffeine
consumption and fetal death, a case-control study in Uruguay. Paediatr. Perinat.
Epidemiol. 20, 100–109.
McElrath, T.F., Hecht, J.L., Dammann, O., Boggess, K., Onderdonk, A., Markenson,
G., Harper, M., Delpapa, E., Allred, E.N., Leviton, A.ELGAN Study Investigators,
2008. Pregnancy disorders that lead to delivery before the 28th week of
gestation, an epidemiologic approach to classification. Am. J. Epidemiol. 168,
980–989.
Mikkelsen, T.B., Osterdal, M.L., Knudsen, V.K., Haugen, M., Meltzer, H.M., Bakketeig,
L., Olsen, S.F., 2008. Association between a Mediterranean-type diet and risk of
preterm birth among Danish women, a prospective cohort study. Acta Obstet.
Gynecol. Scand. 87, 325–330.
Miller, E.A., Manning, S.E., Rasmussen, S.A., Reefhuis, J., Honein, M.A.The National
Birth Defects Prevention Study, 2009. Maternal exposure to tobacco smoke,
alcohol and caffeine, and risk of anorectal atresia, National Birth Defects
Prevention Study 1997–2003. Paediatr. Perinat. Epidemiol. 23, 9–17.
2575
Mongraw-Chaffin, M.L., Cohn, B.A., Cohen, R.D., Christianson, R.E., 2008. Maternal
smoking, alcohol consumption, and caffeine consumption during pregnancy in
relation to a son’s risk of persistent cryptorchidism, a prospective study in the
Child Health and Development Studies cohort, 1959–1967. Am. J. Epidemiol.
167, 257–261.
Morrison, A.S., 1984. Control of cigarette smoking in evaluating the association of
coffee drinking and bladder cancer. In: MacMahon, B., Subimura, T. (Eds.),
Banbury Report 17; Coffee and Health. Cold Spring Harbor Laboratory, USA, pp.
127–134.
Naeye, R.L., 1987. Do placental weights have clinical significance? Hum. Pathol. 18,
387–391.
Natsume, N., Kawai, T., Ogi, N., Yoshida, W., 2000. Maternal risk factors in cleft lip
and palate, case control study. Br. J. Oral Maxillofac. Surg. 38, 23–25.
Nebert, D.W., McKinnon, R.A., Puga, A., 1996. Human drug-metabolizing enzyme
polymorphisms, effects on risk of toxicity and cancer. DNA Cell Biol. 15, 273–
280.
Nordmark, A., Lundgren, S., Cnattingius, S., Rane, A., 1999. Dietary caffeine as a
probe agent for assessment of cytochromone P4501A2 activity in random urine
samples. Br. J. Clin. Pharmacol. 47, 397–402.
Nussey, S.S., Whitehead, S.A., 2001. Endocrinology, An Integrated Approach. Taylor
and Francis, Oxford, UK.
Okon, M.A., Laird, S.M., Tuckerman, E.M., Li, T.C., 1998. Serum androgen levels in
women who have recurrent miscarriages and their correlation with markers of
endometrial function. Fertil. Steril. 69, 682–690.
Ørskou, J., Henriksen, T.B., Kesmodel, U., Secher, N.J., 2003. Maternal characteristics
and lifestyle factors and the risk of delivering high birth weight infants. Obstet.
Gynecol. 102, 115–120.
Ott, W.J., 2006. Sonographic diagnosis of fetal growth restriction. Clin. Obstet.
Gynecol. 49, 295–307.
Parazzini, F., Chiaffarino, F., Chatenoud, L., Tozzi, L., Cipriani, S., Chiantera, V., Fedele,
L., 2005. Maternal coffee drinking in pregnancy and risk of small for gestational
age birth. Eur. J. Clin. Nutr. 59, 299–301.
Pennell, C.E., Jacobsson, B., Williams, S.M., Buus, R.M., Muglia, L.J., Dolan, S.M.,
Morken, N.H., Ozcelik, H., Lye, S.J. PREBIC Genetics Working Group, Relton, C.,
2007. Genetic epidemiological studies of preterm birth, guidelines for research.
Am. J. Obstet. Gynecol. 196, 107–118.
Piipari, R., Savela, K., Nurminen, T., Hukkanen, J., Raunio, H., Hakkola, J., Mäntylä, T.,
Beaune, P., Edwards, R.J., Boobis, A.R., Anttila, S., 2000. Expression of CYP1A1,
CYP1B1, CYP3A, and polycyclic aromatic hydrocarbon–DNA adduct formation in
bronchoalveolar macrophages of smokers and non-smokers. Int. J. Cancer 86,
610–616.
Ramlau-Hansen, C.H., Thulstrup, A.M., Bonde, J.P., Olsen, J., Bech, B.H., 2008. Semen
quality according to prenatal coffee and present caffeine exposure, two decades
of follow-up of a pregnancy cohort. Hum. Reprod. 23, 2799–2805.
Rasch, V., 2003. Cigarette, alcohol, and caffeine consumption, risk factors for
spontaneous abortion. Acta Obstet. Gynecol. Scand. 82, 182–188.
Redline, R.W., Patterson, P., 1994. Patterns of placental injury correlations with
gestational age, placental weight, and clinical diagnoses. Arch. Pathol. Lab. Med.
118, 698–701.
Rotondi, M., Valenzano, F., Bilancioni, E., Spano, G., Rogonid, M., Giorlandion, C.,
2001. Prenatal measurement of testicular diameter by ultrasonography,
development of fetal male gender and evaluation of testicular descent. Prenat.
Diagn. 21, 112–115.
Sachse, C., Brockmoller, J., Bauer, S., Roots, I., 1999. Functional significance of a C ? a
polymorphism in intron 1 of the cytochromone P450 CYP1A2 gene tested with
caffeine. Br. J. Clin. Pharmacol. 47, 445–449.
Saltvedt, S., Almström, H., Kublickas, M., Reilly, M., Valentin, L., Grunewald, C., 2004.
Ultrasound dating at 12–14 or 15–20 weeks of gestation? A prospective crossvalidation of established dating formulae in a population of in-vitro fertilized
pregnancies randomized to early or late dating scan. Ultrasound Obstet.
Gynecol. 24, 42–50.
Santos, I.S., Victora, C.G., Huttly, S., Carvalhal, J.B., 1998. Caffeine intake and low
birth weight, a population-based case-control study. Am. J. Epidemiol. 147,
620–627.
Santos, I.S., Matijasevich, A., Valle, N.C.J., 2005. Mate drinking during pregnancy and
risk of preterm and small for gestational age birth. J. Nutr. 135, 1120–1123.
Sata, F., Yamada, H., Suzuki, K., Saijo, Y., Kato, E.H., Morikawa, M., Minakami, H.,
Kishi, R., 2005. Caffeine intake, CYP1A2 polymorphism and the risk of recurrent
pregnancy loss. Mol. Hum. Reprod. 11, 357–360.
Savitz, D.A., 2008a. Disaggregating preterm birth to determine etiology. Am. J.
Epidemiol. 68, 990–992.
Savitz, D.A., Blackmore, C.A., Thorp, J.M., 1991. Epidemiology of preterm delivery,
etiologic heterogeneity. Am. J. Obstet. Gynecol. 164, 467–471.
Savitz, D.A., Hertz-Picciotto, I., Poole, C., Olsham, A.F., 2002. Epidemiologic measures
of the course and outcome of pregnancy. Epidemiol. Rev. 24, 91–101.
Savitz, D.A., Dole, N., Herring, A.M., Kaczor, D., Murphy, J., Siega-Riz, A.M., Thorp Jr.,
J.M., MacDonald, T.L., 2005. Should spontaneous and medically indicated
preterm births be separated for studying aetiology? Paediatr. Perinat.
Savitz, D.A., Chan, R.L., Herring, A.H., Howards, P.P., Hartmann, K.E., 2008b. Caffeine
and miscarriage risk. Epidemiology 19, 55–62.
Schlesselman, J.J., 1982. Case Control Studies Design Conduct, Analysis. Oxford
University Press, New York. pp. 135–136.
Schmid, T.E., Eskenazi, B., Baumgartner, A., Marchetti, F., Young, S., Weldon, R.,
Anderson, D., Wyrobek, A.J., 2007. The effects of male age on sperm DNA
damage in healthy non-smokers. Hum. Reprod. 22, 180–187.
2576
Schmidt, R.J., Romitti, P.A., Burns, T.L., Browne, M.L., Druschel, C.M., Olney, R.S., and
the National Birth Defects Prevention Study, 2009. Maternal caffeine
consumption and risk of neural tube defects. Birth Defects Res. A Clin. Mol.
Teratol. 85, 879–889.
Schreiber, G.B., Robins, M., Maffeo, C.E., Masters, M.N., Bond, A.P., Morganstein, D.,
1988. Confounders contributing to the reported associations of coffee or
caffeine with disease. Prev. Med. 17, 295–309.
Shimada, T., Watanabe, J., Kawajiri, K., Sutter, T.R., Guengerich, F.P., Gillam, E.M.,
Inoue, K., 1999. Catalytic properties of polymorphic human cytochrome P450
1B1 variants. Carcinogenesis 20, 1607–1613.
Signorello, L.B., McLaughlin, J.K., 2004. Maternal caffeine consumption and
spontaneous abortion, a review of the epidemiologic evidence. Epidemiology
15, 229–239.
Signorello, L.B., Nordmark, A., Granath, F., Blot, W.J., McLaughlin, J.K., Annerén, G.,
Lundgren, S., Ekbom, A., Rane, A., Cnattingius, S., 2001. Caffeine metabolism and
the risk of spontaneous abortion of normal karyotype fetuses. Obstet. Gynecol.
98, 1059–1066.
Simpson, J.L., 1990. Incidence and timing of pregnancy losses, relevance to
evaluating safety and early prenatal diagnosis. Am. J. Med. Genet. 35, 165–173.
Slickers, J.E., Olshan, A.F., Siega-Riz, A.M., Honein, M.A., Aylsworth, A.S., for the
National Birth Defects Prevention Study, 2008. Maternal body mass index and
lifestyle exposures and the risk of bilateral renal agenesis or hypoplasia: The
National Birth Defects Prevention Study. Am. J. Epidemiol. 168, 1259–1267.
Smith, G.C., 2004. First trimester origins of fetal growth impairment. Semin.
Perinatol. 28, 41–50.
Smith, G.C., Smith, M.F., McNay, M.B., Fleming, J.E., 1998. First-trimester growth and
the risk of low birth weight. N. Engl. J. Med. 339, 1817–1822.
Sobreiro, B.P., Lucon, A.M., Pasqualotto, F.F., Hallak, J., Athayde, K.S., Arap, S., 2005.
Semen analysis in fertile patients undergoing vasectomy, reference values and
variations according to age, length of sexual abstinence, seasonality, smoking
habits and caffeine intake. Sao Paulo Med. J. 123, 161–166.
Spira, A., 1998. The use of fecundability in epidemiological surveys. Hum. Reprod.
13, 1753–1756.
Stein, Z., Susser, M., 1991. Miscarriage, caffeine and the epiphenomena of
pregnancy, the causal model. Epidemiology 2, 163–167.
Tan, P.C., Jacob, R., Quek, K.F., Omar, S.Z., 2006. The fetal sex ratio and metabolic,
biochemical, haematological and clinical indicators of severity of hyperemesis
gravidarum. Br. J. Obstet. Gynaecol. 113, 733–737.
Thame, M., Osmond, C., Bennett, F., Wilks, R., Forrester, T., 2004. Fetal growth is
directly related to maternal anthropometry and placental volume. Eur. J. Clin.
Nutr. 58, 894–900.
Tingen, C., Stanford, J.B., Dunson, D.B., 2004. Methodologic and statistical
approaches to studying human fertility and environmental exposure. Environ.
Health Perspect. 112, 87–93.
Tolstrup, J.S., Kjaer, S.K., Munk, C., Madsen, L.B., Ottesen, B., Bergholt, T.,
Grønbaek, M., 2003. Does caffeine and alcohol intake before pregnancy
predict the occurrence of spontaneous abortion? Hum. Reprod. 18, 2704–
2710.
Torfs, C.P., Christianson, R.E., 2000. Effect of maternal smoking and coffee
consumption on the risk of having a recognized Down syndrome pregnancy.
Am. J. Epidemiol. 152, 1185–1191.
Tough, S.C., Newburn-Cook, C.V., White, D.E., Fraser-Lee, M.J., Faber, A.J., Frick, C.,
Svenson, L.W., Sauve, R., 2003. Do maternal characteristics and past pregnancy
experiences predict preterm delivery among women aged 20–34? J. Obstet.
Gynaecol. Can. 25, 656–666.
Tsubouchi, H., Shimoya, K., Hayashi, S., Toda, M., Morimoto, K., Murata, Y., 2006.
Effect of coffee intake on blood flow and maternal stress during the third
trimester of pregnancy. Int. J. Gynaecol. Obstet. 92, 19–22.
Vik, T., Vatten, L., Markestad, T., Jacobsen, G., Bakketeig, L.S., 1997. Dubowitz
assessment of gestational age and agreement with prenatal methods. Am. J.
Perinatol. 14, 369–373.
Vik, T., Bakketeig, L.S., Trygg, K.U., Lund-Larsen, K., Jacobsen, G., 2003. High caffeine
consumption in the third trimester of pregnancy, gender-specific effects on fetal
growth. Paediatr. Perinat. Epidemiol. 17, 324–331.
Virtanen, H.E., Tapanainen, A.E., Kaleva, M.M., Suomi, A.M., Main, K.M., Skakkebaek,
M.E., Toppari, J., 2006. Mild gestational diabetes as a risk factor for congenital
cryptorchidism. J. Clin. Endocrinol. Metab. 91, 4862–4865.
Weigel, R.M., Weigel, M.M., 1989. Nausea and vomiting of early pregnancy and
pregnancy outcome, a meta-analytical review. Br. J. Obstet. Gynaecol. 96, 1312–
1318.
Weigel, M.M., Reyes, M., Caiza, M.E., Tello, N., Castro, N.P., Cespededs, S., Duchicela,
S., Betancourt, M., 2006. In the nausea and vomiting of early pregnancy really
feto-protective? J. Perinat. Med. 34, 115–122.
Weinberg, C.R., Hertz-Picciotto, I., Baird, D.D., Wilcox, A.J., 1992. Efficiency and bias
in studies of early pregnancy loss. Epidemiology 3, 17–22.
Wen, W., Xiao, U.S., Jacobs, D.R., Brown, J.E., 2001. The associations of maternal
caffeine consumption and nausea with spontaneous abortion. Epidemiology 12,
38–42.
Weng, X., Odouli, R., Li, D.K., 2008. Maternal caffeine consumption during
pregnancy and the risk of miscarriage, a prospective cohort study. Am. J.
Obstet. Gynecol. 198, 279–286.
Wilcox, A.J., Baird, D.D., Weinberg, C.R., 1999. Time of implantation of the conceptus
and loss of pregnancy. N. Engl. J. Med. 340, 1796–1799.
Wisborg, K., Kesmodel, U., Bech, B.H., Hedegaard, M., Henriksen, T.B., 2003. Maternal
consumption of coffee during pregnancy and stillbirth and infant death in first
year of life, prospective study. BMJ 326, 420–423.
Xue, F., Willett, W.C., Rosner, B.A., Forman, M.R., Michels, K.B., 2008. Parental
characteristics as predictors of birthweight. Hum. Reprod. 23, 168–177.
Yoon, H., Chung, I.S., Seol, E.Y., Park, B.Y., Park, H.W., 2000. Development of the lip
and palate in staged human embryos and early fetuses. Yonsei Med. J. 41, 477–
484.
Yusuf, R.Z., Naeem, R., 2004. Cytogenic abnormalities in products of conception, a
relationship revisited. Am. J. Reprod. Immunol. 52, 88–96.
Zavela, K.J., Barnett, J.E., Smedi, K.J., Istvan, J.A., Matarazzo, J.D., 1990. Concurrent
use of cigarettes, alcohol and coffee. J. Appl. Soc. Psychol. 20, 835–845.
Zhang, X., Platt, R.W., Cnattingius, S., Joseph, K.S., Kramer, M.S., 2007. The use of
customised versus population-based birthweight standards in predicting
perinatal mortality. Br. J. Obstet. Gynaecol. 114, 474–477.
Zusterzeel, P.L., Nelen, W.L., Roelofs, H.M., Peters, W.H., Blom, H.J., Steegers, E.A.,
2000. Polymorphisms in biotransformation enzymes and the risk for recurrent
early pregnancy loss. Mol. Hum. Reprod. 6, 474–478.
Emergency Department Visits
Involving Energy Drinks and
Limitations of the Drug Abuse
Warning Network (DAWN)
Prepared for the American Beverage
Association by PinneyAssociates
July 25, 2013
Page 1 of 18
PinneyAssociates, Inc.
Table of Contents
1
Executive Summary ................................................................................................. 3
2
Drug Abuse Warning Network (DAWN) ................................................................... 3
3
Data Analysis Approach ........................................................................................... 4
4 Increasing Number of Energy Drink-Related ED Visits: Real Phenomenon or
Artifact? ........................................................................................................................... 4
4.1
Limitations of DAWN .......................................................................................... 9
4.1.1
U.S.
Representativeness of the Sample and Validity of Projected Rates for the
9
4.1.2
Reliability of Self-Reported Data................................................................ 10
4.1.3
Inability to Determine Causation ................................................................ 10
5
Potential Issues ...................................................................................................... 11
6
Conclusion ............................................................................................................. 13
8
Appendix ................................................................................................................ 14
Page 2 of 18
1 Executive Summary
The Substance Abuse and Mental Health Services Administration (SAMHSA) released
a report in January 2013, based on data from the Drug Abuse Warning Network
(DAWN), suggesting an increase in the number of emergency department (ED) visits
involving energy drinks and concluding that the consumption of energy drinks is a “rising
public health problem”. At the request of the American Beverage Association, Pinney
Associates (PA) was asked to conduct a review of the DAWN report and its findings.
Overall, reports of energy drink-related ED visits need to be viewed in a broader
context, as an analysis of DAWN public use data indicates that drug-related ED visits
have also increased (both by a similar proportion and absolute magnitude as compared
to energy drinks) for a number of other products, including infant formula, vitamins, and
laxatives. Furthermore, the vast majority of energy drink-related ED visits appear to
have been occasioned by non-serious medical conditions: 84.4% of visits related to
caffeine/multivitamins resulted in discharge home, rather than admission to a treatment
facility. In comparison, only 75.5% of alternative medicine-related ED visits resulted in
home discharge. Given that there are a number of other products demonstrating
comparable increases in ED visits, and that these products appear to be associated
with a less benign profile than that associated with energy drinks, it is unclear why
energy drinks have been singled out by SAMHSA as a public health concern. The
DAWN public use data do not support the public health concern flagged by SAMSHA.
2 Drug Abuse Warning Network (DAWN)
DAWN is a public health surveillance system that monitors “drug-related” visits to
hospital EDs. Each year DAWN produces estimates of such visits for the nation as a
whole and for selected metropolitan areas. To be a DAWN case, the ED visit must
involve a drug, either as the direct cause of the visit or as a contributing factor. Such a
visit is referred to as a “drug related visit.” The reason a patient used a drug is not part
of the criteria for considering a visit to be drug-related. Drugs include: alcohol1; illegal
drugs, such as cocaine, heroin, and marijuana; pharmaceuticals (e.g., over-the-counter
medicines and prescription medications); and nutraceuticals, such as nutritional
supplements, vitamins, and caffeine-containing products. DAWN cases are identified
by the systematic review of ED medical records in participating hospitals. DAWN cases
broadly encompass all types of drug-related events, including accidental ingestion and
adverse reactions, as well as explicit drug abuse. SAMHSA noted in its report on
energy drinks that although energy drinks are not treated as drugs by the FDA, ED visits
involving energy drinks were classified as adverse reactions if the chart documented
them as such. 2
1
Alcohol is considered a reportable drug when consumed by patients aged 20 or younger. For patients
aged 21 and older, alcohol is reported only when it is used in conjunction with other drugs.
2
Within DAWN, an ED visit is categorized as an adverse reaction when the chart documents that a
prescription or over-the-counter pharmaceutical, taken as prescribed or directed, produced an adverse
drug reaction, side effect, drug-drug interaction, or drug-alcohol interaction.
Page 3 of 18
The exact DAWN survey methodology has been adjusted over time in order to,
according to SAMHSA, “improve the quality, reliability, and generalizability of the
information produced by DAWN” (Source: DAWN 2010 Codebook). The current
approach, which was developed based on recommendations from a 1997 panel of
experts and a 2-year SAMHSA evaluation of design alternatives, was introduced in
2003, but not fully implemented until the 2004 data collection year.
3 Data Analysis Approach
In order to put the SAMHSA findings on energy drinks into perspective, PA conducted a
number of additional analyses using the DAWN public-use dataset. However, there is
an important caveat to these analyses that must be acknowledged; namely, information
on the use of energy drinks per se is not currently available in the public-use data file.
Rather, the public-use data file only contains information on the larger category of
“caffeine/multivitamins,” of which the “energy drinks” category is a subset. As this larger
category appears to be mostly comprised of energy drink-related visits (about 80%
overall, from 2005-2011) information pertaining to caffeine/multivitamin-related ED visits
are used as a proxy for energy drink-related visits in all reported analyses. Outreach to
SAMHSA revealed that the agency has received several requests for the specific
energy drink data, but thus far has declined to make these data public.
4 Increasing Number of Energy Drink-Related ED Visits: Real
Phenomenon or Artifact?
According to the SAMHSA report, the number of ED visits involving energy drinks
doubled from 10,068 visits in 2007 to 20,783 visits in 2011. 3 Notably, however, an
analysis of DAWN public-use data indicates that the total number of overall drug-related
ED visits (regardless of the specific drug/s involved) also increased between 2007 and
2011, rising from 3.9 million visits to 5.1 million visits. Therefore, the increase in energy
drink-related visits should be understood in the context of an increase in overall drugrelated ED visits. It is not known whether this reflects a real increase in the utilization of
EDs, or an artifact perhaps resulting from change in the data collection or case
identification methodology. In 2007, energy drink-related visits comprised 0.25% of all
drug-related ED visits. In 2011, energy drink-related visits comprised 0.41% of all drugrelated ED visits.
Furthermore, as shown in Table 1 below, estimated drug-related ED visits appear to
have increased not only for energy drinks, but for a number of other drugs/products,
including infant formula, alternative medications, and other miscellaneous products such
as dermatological agents (e.g., Vick’s, hand lotion), gastrointestinal agents (e.g.,
laxatives), isopropyl (rubbing) alcohol, and ophthalmic preparations (e.g., eye drops,
contact solution). Not only have drug-related ED visits increased for these other
products by similar proportions as for energy drinks, for many, their absolute magnitude
is similar, too (see Figure 1 below). In addition, energy drink-related ED visits appear to
3
It is important to note that these are not raw numbers of visits, but estimates projected to a national
sample. The limitations of the weighting system used to derive these projected estimated are discussed
in Section 4.1.1 below.
Page 4 of 18
be more likely to be associated with non-serious complaints that do not require further
medical follow-up, compared to ED visits related to other product/medications. Yet,
increasing ED visits associated with these other products have not been identified as a
public health concern.
Figure 1 Number of ED Visits Related to Specific Products
40000
2007
2011
35000
Number of ED Visits
30000
25000
20000
15000
10000
5000
0
Energy Drinks
Infant Formula
Vitamins
Product
Laxatives
It is unclear whether these data reflect an increase in the levels of accidental and/or
intentional exposure to substances and drugs in general, including energy drinks, or if
there are methodological and statistical processes that may give the appearance of
notable increases in drug-related ED visits. It is possible, for example, that the
observed increases in some categories could be due to increased awareness by health
professionals of certain substances, or increased perception of certain categories as
problematic. This could lead to either increased detection of such substances (e.g., if
the medical interviewer asks about them more than previously) or increased attribution
of ED visits to the substance (e.g., if the medical interviewer is more likely to record the
substance or to name it as a factor in the ED visit).
Page 5 of 18
Table 1 Number of ED Visits Related to Specific Products
Drug
% Change,
2007-2011
2007
2008
2009
2010
2011
Total drug-related ED visits
3,998,228
4,383,494
4,595,263
4,916,328
5,067,374
26.74%
Total drug reports
6,248,023
6,957,634
7,270,914
7,808,492
8,046,258
28.78%
12,750
18,970
14,415
18,734
29,379
130.42%
10,068
16,059
13,119
15,219
20,783
106.43%
59,389
74,437
80,724
93,749
95,089
60.11%
7,800
8,885
11,020
12,982
12,711
62.96%
11,140
16,364
15,088
16,094
14,946
34.17%
Electrolyte replacement
a
solutions, oral
673
689
855
1,282
1,824
171.03%
Oral nutritional supplements
15,388
15,919
20,835
26,014
33,855
120.01%
12,764
12,019
16,582
22,242
28,212
121.03%
9,499
13,566
13,847
16,369
14,834
56.16%
18,915
26,905
28,857
29,381
29,672
56.87%
Alternative medicines
13,320
15,892
15,951
20,806
24,222
81.85%
Herbal products
8,603
6,661
8,864
11,915
12,508
45.39%
Nutraceutical products
4,385
8,975
7,356
8,600
10,087
130.03%
330
485
128
752
1,760
433.33%
Gastrointestinal agents
78,826
94,468
104,390
101,940
103,358
31.12%
Antidiarrheals
6,947
8,462
8,526
12,113
10,859
56.31%
Caffeine/multivitamin
Energy drinks
Nutritional products
Iron products
Minerals and electrolytes
Infant formula
Vitamin and mineral combinations
Vitamins
Probiotics
Page 6 of 18
Drug
% Change,
2007-2011
2007
2008
2009
2010
2011
19,424
28,053
27,621
29,668
33,861
74.33%
Dermatological agents
30,072
30,438
36,016
44,262
50,632
68.37%
Topical emollients
2,832
2,937
2,972
5,622
4,836
70.76%
Hydrocortisone, topical
2,019
2,817
4,206
4,284
3,997
97.97%
460
1,402
238
1,032
2,204
379.13%
593
471
957
2,361
1,503
153.46%
48,732
53,169
53,652
66,888
93,457
91.78%
6,434
5,930
7,293
8,633
8,936
38.89%
2,252
4,504
2,473
2,779
3,219
42.94%
9,137
9,125
11,828
13,653
14,506
58.76%
Laxatives
Camphor
b
Hydrogen peroxide, topical
Miscellaneous
CNS Stimulants
c
Caffeine
Isopropyl alcohol, topical
Ophthalmic preparations
d
a
Electrolyte replacement solutions include products such as Gatorade, Powerade, Pedialyte, etc.
Camphor includes products such as Vick’s, Biofreeze, etc.
c
Caffeine includes coffee, as well as other caffeine-containing products, including caffeine pills and diet pills.
d
Ophthalmic preparations include contact solution, eye drops, etc.
b
Page 7 of 18
An important consideration in the assessment of drug-related ED visits is the health
outcomes or consequences associated with such visits. While DAWN does not capture
information on the nature of the complaint or symptom severity that prompted the ED
visit, there is information available on the disposition or discharge status of ED visits
that can serve as a proxy for measuring clinical severity and acuity. Table 2 below
shows the results of an analysis of the 2011 DAWN public-use data that was conducted
to determine the percentage of visits resulting in discharge home for all drug-related ED
visits, caffeine/multivitamin-related visits, and for three groups of selected comparator
products (nutritional products, which includes iron products, minerals and electrolytes,
oral nutritional supplements, vitamins; alternative medicines, which includes herbal
products, nutraceutical products, probiotics; and CNS stimulants) (see Appendix Table
5 for additional information on the visit and demographic characteristics associated with
caffeine/multivitamin-related ED visits, as well as the three selected comparator
products).
Of the overall caffeine/multivitamin-related ED visits in 2011, 84.4% resulted in
discharge home. Considering ED visits related to caffeine/multivitamin use only (i.e., no
other drug involvement), the percentage of visits resulting in discharge without any
further follow-up was even higher (88.3%), demonstrating that the vast majority of
energy drink-related ED visits are for non-serious complaints that do not require further
medical care. Notably, home discharge rates for caffeine/multivitamin-related ED visits
are substantially higher than those for drug-related ED visits overall (63.8%). These
findings are consistent with information from the American Association of Poison
Control Centers’ (AAPCC) National Poison Data System which indicates that in cases
involving energy drink exposure where medical outcome was assessed, the vast
majority of cases were considered to be not serious (83% of cases with medical
outcomes classified as “none” or “minor”). 4 This suggests that ED visits associated with
consumption of energy drinks are not as serious as those associated with other drugs.
Table 2 Home discharge rates for selected ED visit types
Visit Type
% of Visits Resulting in
Discharge Home
All drug-related ED visits
63.8%
CNS stimulants-related visits
74.2%
Alternative medicines-related
visits
75.5%
Nutritional products-related visits
80.3%
Caffeine/multivitamin-related
visits
84.4%
4
Bronstein AC, et al. 2011 Annual Report of the American Association of Poison Control Centers’
th
National Poison Data System (NPDS): 29 Annual Report. Clinical Toxicology 2012;50:911-1164. Note:
Energy drinks were added as a generic code to NPDS in 2010. Because only partial year data is
available for 2010, it is not yet possible to assess trends related to energy drinks with these data.
Page 8 of 18
4.1 Limitations of DAWN
Though not directly addressing the reported rise in energy drink-related ED visits, there
are a number of limitations of DAWN that are worth noting.
Representativeness of the Sample and Validity of Projected Rates for the
U.S.
DAWN uses a sample of hospital EDs to estimate national ED visit rates, including 13
major metropolitan areas and a supplementary sample to cover the remainder of the
U.S. In 2002, prior to the most recent DAWN re-design, there were 21 metropolitan
areas included in the sample. The DAWN redesign methodology report called for an
expansion to 48 metropolitan areas in order to provide better national coverage and to
increase the reliability and stability of their estimates. However, in 2004 (the first
complete year of the redesigned DAWN) only 15 metropolitan areas had sufficient
participation to warrant separate, stand-alone estimates. As of 2011 (the latest year for
which public use data are available), the number of metropolitan areas with sufficient
participation was further reduced to 13. Thus, although the expert panel that evaluated
DAWN recommended more participating hospitals to increase reliability, in fact there
are now fewer participating hospitals.
4.1.1
It is important to understand that DAWN’s reporting is not based on a straightforward
enumeration of cases. DAWN projects to a national estimate of cases based on
combining results from two sources: approximately 183 hospitals in 13 major
metropolitan areas, and approximately 50 supplementary hospitals in 2011. Although
the metropolitan hospitals actually report more cases, the supplementary hospitals
actually exert greater influence on the projected national estimate. On average, one
case in the supplementary sample represents 135 weighted cases, whereas one case
in any of the 13 main metropolitan areas represents, on average, fewer than 5 weighted
cases (see Appendix Table 4). Therefore, a single case from a supplementary hospital
can count 27 times more than a case from one of the metropolitan hospitals that report
data to DAWN. This can distort the estimate. For example, a small ‘outbreak’ at a
community hospital could potentially skew the national statistics; a single case of energy
drink use presenting to a hospital in the supplementary sample could be counted as
though it were 863 cases (the maximum weight for a single case in 2011), possibly
seriously skewing the national statistics and resulting in misleading trend data.
In 2011, the vast majority (85.6%) of weighted caffeine/multivitamin-related ED visits
were derived from the supplementary sample. This does not appear to be unique to
caffeine/multivitamins, however, as an analysis of selected comparator products (i.e.,
nutritional products, alternative medicines, and CNS stimulants) revealed that for these
three other drug classes/product categories the bulk of the weighted reporting is also
coming from the supplementary sample: 83.7% for nutritional products, 83.4% for
alternative medicines, and 87.3% for CNS stimulants.
Using the publicly available DAWN data, we examined trends in caffeine/multivitaminrelated ED visits by individual metropolitan area and observed a variable pattern.
Among the 11 metropolitan areas with available data between 2007-2011, two areas
experienced a decrease in caffeine/multivitamin-related ED visits during this time period
Page 9 of 18
(Denver, Phoenix); four areas experienced an increase between 50-100% (Boston,
Chicago, Houston, Minneapolis-St. Paul); and five areas (Dade County (Miami), Detroit,
New York City, San Francisco, and Seattle) experienced an increase greater than
100%. This may imply that there are regional variations in trends in ED visits related to
energy drinks or that there are regional variations in the characterization of ED visits,
possibly from a greater local awareness in the higher reporting areas. An analysis of
selected comparator products also revealed regional variation in ED visits. For the
category of CNS stimulants, for example, one metropolitan area experienced a
decrease in ED-related visits between 2007 and 2011; one area experienced an
increase of less than 50%; five areas experienced an increase between 50-100% and
two areas experienced an increase greater than 100%.
Reliability of Self-Reported Data
The reliability of DAWN data is dependent on information listed by the provider on the
ED medical chart, which is typically based on patient self-report taken by the triage
nurse. Therefore, the drugs actually involved in ED visits might not all be identified and
documented. As noted in the SAMHSA report, of the 20,783 ED visits involving energy
drinks in 2011, more than half (58%) were reported to involve energy drinks only.
However, it is possible that while some patients presenting to the ED may have readily
reported use of an energy drink (a legal product, and thus more likely to be considered
socially acceptable), they may have been reluctant to report any other drug use that
may have occurred in conjunction with their use of an energy drink (e.g., use of illegal
drugs, drugs for which there was no valid prescription or use of alcohol by those under
legal age). Further, as described above, the salience of certain drugs/substances and
the perception of the drug/substance as a problem could also affect reporting by the
provider.
4.1.2
Inability to Determine Causation
Many drug-related ED visits involve multiple drugs. As noted in the SAMHSA report, of
the 20,783 ED visits involving energy drinks in 2011, 42% reportedly involved other
drugs. Use of pharmaceuticals was most commonly reported in conjunction with energy
drink use (27%), with 9% of visits involving energy drinks and central nervous
stimulants. About 13% of visits involved energy drinks and alcohol and 10% of visits
involved energy drinks and illicit drugs, with 5% involving energy drinks and marijuana.
In these instances, it may be difficult or impossible to determine whether a single drug
or product is responsible for the visit or if the visit was the result of the interaction
between the drugs. Furthermore, important information that could aid in assessing
causation is not captured (e.g., nature of the complaint/symptoms that brought the
patient to the ED, overall health of the patient, amount used/exposure information).
Importantly, there is no specific information on consumption of other caffeine-containing
products (e.g., coffee – which is included in the larger caffeine category by DAWN, but
not listed as a specific product). This is particularly important given the wide variability
in caffeine content of popular brands of coffee. According to an analysis prepared for
4.1.3
Page 10 of 18
the Food and Drug Administration (FDA) on caffeine consumption in the U.S. 5, the
mean amount of caffeine consumed by the U.S. population has remained relatively
stable between 2003 and 2008 at approximately 300 milligrams per person per day
despite the entry of energy drinks into the marketplace. Furthermore, according to the
same analysis, energy drinks contribute a small portion of the caffeine consumed, with
major sources of caffeine being coffee, soft drinks and tea.
5 Potential Issues
The estimates provided in the SAMHSA report are based solely on number of ED visits,
and do not account for the availability of the product (i.e., sales). As shown in Table 3
(which includes data for the years 2007-2011, since as noted by SAMHSA, statistical
tests were not used until 2007 when the number of ED visits involving energy drinks
exceeded 10,000) and Figure 2 (which displays data for the years 2005-2011,
consistent with the figure presented in the SAMHSA report), the increase in energy
drink-related ED visits was accompanied by an increase in the number of cases of
energy drinks sold. However, ED visits still appear to be increasing at a higher rate
than sales.
Table 3 Energy drink-related ED visits and number of cases of energy drinks sold (2007-2011)
2007
2008
2009
2010
2011
% Change
2007-2011
Number of
energy
drinkrelated
visits
10,068
16,059
13,119
15,219
20,783
106.4%
Cases sold
±
(millions)
234.1
244.5
240.1
261.5
305.0
30.3%
Number of
energydrink
related
visits per 1
million
cases sold
43.0
65.7
54.6
58.2
68.1
58.4%
±
Source: Beverage Digest Fact Book
5
Caffeine Intake by the U.S. Population. Prepared by Laszlo P. Somogyi, Ph.D. for the Food and Drug
Administration, Oakridge National Laboratory. Subcontract Number 70000073494. Completed
September 2009 and revised August 2010. Available at:
http://www.fda.gov/downloads/AboutFDA/CentersOffices/OfficeofFoods/CFSAN/CFSANFOIAElectronicR
eadingRoom/UCM333191.pdf
Page 11 of 18
Figure 2 Energy drink-related ED visits and cases of energy drinks sold (in millions), 2005-2011
Page 12 of 18
6 Conclusion
Although the DAWN report has attracted a lot of attention, careful analysis of the report
and the public data underlying it, do not appear to be consistent with a signal of
substantial medical harm. The vast majority of caffeine/multivitamin-related ED visits
appear to be associated with non-serious complaints that do not require further medical
follow-up, as 84.4% of visits related to these products resulted in discharge home, a
higher rate than observed for other products. The reported rate of ED visits related to
caffeine/multivitamins remains quite small, representing a tiny fraction of the overall
visits to EDs each year. Finally, the limitations of the DAWN system suggest caution in
basing public health policy on the results relative to energy drinks.
Page 13 of 18
8 Appendix
Table 4 DAWN weighting by metro area (2011)
Number of
Cases,
Unweighted
BOSTON-CAMBRIDGE-QUINCY,MANHMSA:(1)
NEW YORK CITY - 5 BUROUGHS (PART OF
NEW YORK- NEWARK-EDISON, NY-NJ-PA
MSA):(2)
CHICAGO-NAPERVILLE-JOLIET, IL-IN-WI
MSA:(3)
DETROIT-WARREN-LIVONIA, MI MSA:(4)
MINNEAPOLIS-ST. PAUL-BLOOMINGTON,
MN-WI MSA:(5)
FORT LAUDERALE DIVISION OF MIAMIFORT LAUDERDALE, FL MSA:(6)
DADE COUNTY DIVISION OF MIAMI-FORT
LAUDERDALE, FL MSA:(7)
HOUSTON-BAYTOWN-SUGAR LAND, TX
MSA:(8)
DENVER-AURORA, CO MSA:(9)
PHOENIX-MESA-SCOTTSDALE, AZ
MSA:(10)
OAKLAND DIVISION OF SAN FRANCISCOOAKLAND-FREMONT, CA MSA:(11)
% of
Unweighted
Cases
Average
Weight
Minimum
Weight
Maximum
Weight
24,889
10.86%
3.86
1.60
8.54
39,776
17.35%
3.13
0.94
22.84
21,918
22,502
9.56%
9.82%
6.68
4.20
1.42
1.23
28.77
11.62
12,049
5.26%
4.50
1.33
8.04
5,352
2.33%
6.15
2.59
14.30
7,101
3.10%
4.46
2.57
8.57
9,115
12,112
3.98%
5.28%
10.31
3.01
3.32
1.10
27.90
7.34
13,166
5.74%
4.76
1.05
15.87
2,462
1.07%
13.29
9.22
18.18
CONFIDENTIAL
CONFIDENTIAL
SAN FRANCISCO DIVISION OF SAN
FRANCISCO-OAKLAND-FREMONT, CA
MSA:(12)
SEATTLE-TACOMA-BELLEVUE, WA
MSA:(13)
ALL OTHER LOCATIONS:(14) (a.k.a.
"supplementary sample")
8,936
3.90%
4.09
1.14
10.06
18,973
8.28%
2.86
1.03
7.74
30,860
13.46%
135.13
2.01
862.82
Page 15 of 18
Table 5 Visit characteristics and demographics for caffeine/multivitamin-related
ED visits, nutritional products-related ED visits, alternative medicine-related ED
visits and CNS stimulant-related ED visits (2011)
Nutritional
Products
95,089
Alternative
Medicines
24,222
CNS Stimulants
Total ED Visits
Caffeine/Multivitamin
Products
29,379
Combinations
Product Only
14,393 (48.99%)
63,780 (67.07%)
11,374 (46.96%)
45,951 (49.17%)
11,952 (40.68%)
11,090 (11.66%)
4,497 (18.57%)
40,648 (43.49%)
8,615 (29.32%)
1,644 (1.73%)
1,523 (6.29%)
17,118 (18.32%)
3,701 (12.60%)
201 (0.21%)
1,653 (6.82%)
12,914 (13.82%)
3,503 (11.92%)
23,735 (24.96%)
8,870 (36.62%)
14,974 (16.02%)
Quarter
First Quarter
5,580 (18.99%)
25,279 (26.59%)
9,059 (37.40%)
20,909 (22.37%)
Second Quarter
7,764 (26.43%)
26,784 (28.17%)
5,738 (23.69%)
25,739 (27.54%)
Third Quarter
8,503 (28.94%)
22,483 (23.64%)
5,485 (22.64%)
26,334 (28.18%)
Fourth Quarter
7,532 (25.64%)
20,542 (21.60%)
3,939 (16.26%)
20,475 (21.91%)
6,367 (21.67%)
14,965 (15.74%)
3,605 (14.88%)
16,914 (18.10%)
5,044 (17.17%)
18,738 (19.71%)
4,274 (17.64%)
18,896 (20.22%)
8,236 (28.03%)
29,750 (31.29%)
9,610 (39.68%)
27,655 (29.59%)
9,733 (33.13%)
31,637 (33.27%)
6,734 (27.80%)
29,993 (32.09%)
14,393 (48.99%)
63,780 (67.07%)
11,374 (46.96%)
45,951 (49.17%)
Product, Any
Pharmaceutical
Combination
Product, Any Alcohol
Combination
Product, Any Illicit
Drug Combination
Product, 2+
Substances, Not
Misuse/Abuse
93,457
Visit Characteristics
Part of the Day
Early morning
(12:00-5:59 AM)
Morning (6:00-11:59
AM)
Afternoon (12:005:59 PM)
Evening/Night (6:0011:59 PM)
Number of
Substances
One
CONFIDENTIAL
CONFIDENTIAL
Nutritional
Products
31,308 (32.93%)
Alternative
Medicines
12,848 (53.04%)
CNS Stimulants
Two or more
Products
14,986 (51.01%)
Case Type
Suicide Attempt
917 (3.12%)
1,473 (1.55%)
1,363 (5.63%)
4,715 (5.05%)
Seeking Detox
364 (1.24%)
5 (0.01%)
14 (0.06%)
2,272 (2.43%)
Alcohol Only
(Age<21)
Adverse Reaction
0 (0.00%)
0 (0.00%)
0 (0.00%)
0 (0.00%)
15,914 (54.17%)
79,638 (83.75%)
16,656 (68.76%)
41,311 (44.20%)
13,061 (44.46%)
57,447 (60.41%)
8,528 (35.21%)
28,970 (31.00%)
Product, Any
Pharmaceutical
Combination
Product, Any
Alcohol Combination
Drug Combination
Product, 2+
Substances, Not
Misuse/Abuse
Overmedication
0 (0.00%)
0 (0.00%)
0 (0.00%)
0 (0.00%)
0 (0.00%)
820 (0.86%)
659 (2.72%)
1,594 (1.71%)
5 (0.02%)
5 (0.00%)
0 (0.00%)
5 (0.00%)
2,849 (9.70%)
21,366 (22.47%)
7,469 (30.84%)
10,743 (11.49%)
1,247 (4.25%)
9,240 (9.72%)
1,769 (7.30%)
10,959 (11.73%)
Malicious Poisoning
30 (0.10%)
293 (0.31%)
0 (0.00%)
94 (0.10%)
Accidental Ingestion
232 (0.79%)
2,883 (3.03%)
1,693 (6.99%)
4,510 (4.83%)
Other
10,675 (36.34%)
1,557 (1.64%)
2,729 (11.27%)
29,596 (31.67%)
Disposition
Discharged Home
24,798 (84.41%)
76,326 (80.27%)
18,295 (75.53%)
69,379 (74.24%)
Product Only
12,714 (43.28%)
58,968 (62.01%)
9,470 (39.09%)
39,000 (41.73%)
9,722 (33.09%)
6,949 (7.31%)
2,613 (10.79%)
27,820 (29.77%)
6,416 (21.84%)
461 (0.48%)
1,060 (4.37%)
11,016 (11.79%)
3,103 (10.56%)
101 (0.11%)
767 (3.17%)
7,032 (7.52%)
3,431 (11.68%)
14,007 (14.73%)
6,545 (27.02%)
10,506 (11.24%)
15 (0.05%)
100 (0.11%)
8 (0.03%)
260 (0.28%)
363 (1.24%)
430 (0.45%)
32 (0.13%)
2,134 (2.28%)
367 (1.25%)
1,133 (1.19%)
288 (1.19%)
2,074 (2.22%)
Product Only
Product, Any
Pharmaceutical
Combination
Product, Any
Alcohol Combination
Drug Combination
Product, 2+
Substances, Not
Misuse/Abuse
Released to
Police/Jail
Referred to
Detox/Treatment
ICU/Critical Care
47,506 (50.83%)
Page 17 of 18
CONFIDENTIAL
Products
5 (0.02%)
Nutritional
Products
387 (0.41%)
Alternative
Medicines
0 (0.00%)
CNS Stimulants
Chemical
Dependency/Detox,
Psychiatric Unit
Other Inpatient
50 (0.17%)
189 (0.20%)
1,056 (4.36%)
2,973 (3.18%)
1,804 (6.14%)
13,263 (13.95%)
3,653 (15.08%)
5,608 (6.00%)
Transferred
972 (3.31%)
2,244 (2.36%)
697 (2.88%)
9,401 (10.06%)
Left Against Medical
Advice
Died
326 (1.11%)
90 (0.09%)
60 (0.25%)
718 (0.77%)
0 (0.00%)
0 (0.00%)
0 (0.00%)
0 (0.00%)
Other
672 (2.29%)
222 (0.23%)
108 (0.45%)
823 (0.88%)
Not Documented
7 (0.02%)
703 (0.74%)
25 (0.10%)
81 (0.09%)
Sex
Male
20,502 (69.78%)
40,796 (42.90%)
10,684 (44.11%)
54,926 (58.77%)
Female
8,877 (30.22%)
54,293 (57.10%)
13,538 (55.89%)
38,531 (41.23%)
Age Category
0-11
668 (2.27%)
32,032 (33.69%)
2,762 (11.40%)
10,926 (11.69%)
12-17
3,082 (10.49%)
2,345 (2.47%)
1,145 (4.73%)
13,859 (14.83%)
18-24
9,260 (31.52%)
2,627 (2.76%)
3,494 (14.43%)
23,543 (25.19%)
25-34
7,038 (23.96%)
6,510 (6.85%)
4,148 (17.13%)
21,486 (22.99%)
35+
9,332 (31.76%)
51,575 (54.24%)
12,673 (52.32%)
23,643 (25.30%)
Race/Ethnicity
White Only
18,293 (62.26%)
60,953 (64.10%)
17,926 (74.01%)
68,763 (73.58%)
African American
Only
Hispanic or Latino
3,475 (11.83%)
14,800 (15.56%)
2,284 (9.43%)
9,108 (9.75%)
7,055 (24.02%)
16,528 (17.38%)
3,140 (12.96%)
14,404 (15.41%)
All Other Races
556 (1.89%)
2,807 (2.95%)
873 (3.60%)
1,181 (1.26%)
Surgery
5 (0.01%)
Demographics
Page 18 of 18
Original Article
Energy drinks: An emerging public health hazard
for youth
Jennifer L. Pomeranz*', Christina R. Munsell, and Jennifer L. Harris
Rudd center for Food Policy Ec obesit¡ yale universit¡ 3o9 Edwarcls Street,
PO Box 2o8369, Nerv Haven, CT o65zo-8369, USA.
oCorresponding author.
Abstract
Energy drinks are emerging as a public health threat and are
increasingly consumed by youth internationally. Enãrgy drinks contain high levels
of.caffeine, sugar, and novel ingredients, and are oftèn marketed througË youthoriented media and venues. rùfe review these practices and the current inconsisrent
state of labeling. ve also examine international support for regulation of these
products, includ_ing a survey showing that 85 p.. i*t of united stares parenrs
agreed thatregulations requiring caffeine content disclosure and rvarning labels on
energy drinks are warranted. we then examine the regulatory structurelor energy
drinks in the united Stares,_analyzing legal and self-regulatory strategies to prorect
consumers, especially youth, from these potentially dangerous products. Recommended government interventions include revised labeling require-ments, addressing
problematic ingredients, and enacting retail restrictions. \ùüe cãnclude by identifyin!
areas for future research.
Jouryal of Public Health Policy (zory) 34, zS4-z7r. doi:ro.ro57ljphp.zor3.6;
published online 14 March zor3
Keywords: child and adolescent health; energy drinks; marketing; regnlation; law
Introduction
The consumption of sugary beverages is an established public health
concern,_1 with energy di¡nþs emerging as a unique and indäpendent risk
for youth. sales of energy drinks are rising at ã steadyp".ã.'In zorr,
they increased by rz.5 per cent overall, and by r5-3o p.r...rt for the
category leaders, Red Bull and Rockstar.3 In a study of 6oo nationally
advertised beverage products in the united stares, the sale of energy
drinks surpassed that of either sports or fruit drinks.a
'.ç)
uor
j
Macmillan Publishe¡s Lrd . ot97-5897 Joumal of public Health policy yol.
i,q, z,254-z7t
rvrvrv.pa lgrave- journals.com/iphp/
Regulatingenergydrinks
*
The products in this category typically have the word 'energy' in the
product name and contâin high levels of caffeine plus additional ingredients
not found in sodas and juice drinks. (Energy drinks differ from sports drinks
which are marketed to accompany physical activity and contain electrolytes.) The energy drink category includes two types of products: drinks and
shots. Drinks are sold in 8-3zoz. containers. Many are available in large,
non-resealable cans that produce one serving, despite the number of
servings listed on the container.a's Shots come in z-z.5oz. single serving
containers.4 Because there are few data on youth consumption of energy
shots, this article focuses primarily on energy drinks.
A recent study of US high school students revealed that energy drinks
represented 8.8 per cent of sugar-sweetened beverages they consumed,
and more than ro per cent of drinks consumed by males and Hispanic
students.6 Another US study indicated that 3 r per cent of rz-r7 year
olds regularly consume energy drinks.T Similarl¡ a study of German
adolescents found that 53 per cent ried energy drinks and z6 per cent of
adolescents consumed them regularly.s Internationall¡ Thailand was
reported to be the highest per capita consumers of energy drinks in zoo7,
with the United States, Austria, Ireland, New Zealand, Slovenia, and
Kuwait rounding out the top seven countries.e
Energy drink consumption is a potential health hazard for the general
population and especially alarming for youth due to high levels of
caffeine and novel ingredients not normally found in the food supply.to'tt The American Academy of Pediatrics (AAP) stated that'energy
drinks have no place in the diet of children and adolescents' due to their
'stimulant content'r12 but energy drink manufacturers continue to
advertise directly to adolescents in media also viewed by children.l2
A study by the US Department of Health and Human Services revealed
that emergency room (ER) visits involving energy drinks (alone or mixed
with othei substances) increased tenfold from zoo5 to zoo9.r3
The mixing of energy drinks with alcohol is an obvious public health
.orr.errrrt4 but adolescent consumption of energy drinks alone also poses
considerable health risks. Eleven per cent of total ER visits related to
energy drink consumption involved youth aged rz-t7 years 3nd 7 S per
cent of those visits were due to energy drink intake alone." Similarl¡
calls to the Australian poison information center revealed increasing
reports of caffeine toxicity from energy drink consumption among
adolescents. The median age of callers was rT years and more than half
of all calls were due solely to energy drink consumption.ls
:i) ro¡-r Macmill¿nPublishenLtd. o197-5897 Jomalof PublicHealthPolicy
Yol.34,z,Lj4-L7r
255
*
Ponteravetal
The first part of this ârricle builds on previous research about negative
health effects of energy drink consumprion among youth,T'e by discussing the potential health effects of problematic ingredients, inconsisrent
labeling practices, and the marketing of energy drinks to adolescents.
Then it describes international support for increased regulation of energy
drinks; we also report on a survey of US parents that indicates such support to pfotect youth.'We review current regulatory structure for energy
drinks and analyze legal strategies to protect consumers, especially
youth, from these potentially dangerous products. 'We conclude by
identifying areas for future research, in particular the need for more
information about energy shor consumption and its effects.
Inconsistent Labeling
US Food and Drug Administration (FDA) regulations conrain cerrain
requirements for beverage labels but not all manufacturers of energy
drinks designate their products as 'beverages', thus labels are inconsistent across companies. Manufacturers that label energy drinks as
beverages comply with the Nutrition Labeling and Education Act of
r99o (NLEA). Others mislabel their producs as dietary supplements
and comply with labeling required by the Dietary Supplement Health
and Education Act of ry94 (DSHEA). However, DSHEA has significantly more lax requirements and manufacturers can list ingredients on
supplement facts panels that would nor be permitted under the NLEA.16
If there are no macronutrients in a product, manufacturers of dietary
supplements can eliminate disclosure of the macronutrienr list on rhe
supplements factpanel, unlike beverage manufacrurers who must list rhe
amount as tero.77
The Food, Drug, and Cosmetic Act (FDCA) does not require caffeine
disclosure for beverages or supplements. American Beverage Association
(ABA) member companies and some independent ones disclose caffeine
voluntaril¡18 but as many manufactur..r ào not, consumers would have
to call these companies directly to obtain information about the caffeine
content.
Ingredients and Health Risks
Energy drinks are generally composed of sugar andlor artificial sweereners, caffeine, and additional ingredients, many of them in high
256
rQ zot3 Macmillan Publishers
Ltd. ot97-5897 Journal of Public Health Policy Yol. 34 z, 254-z7t
*
quantities or novel for beverages, such as guârana and taurine. Under the
FDCA, ingredients added to beverages are considered food additives,
and must be pre-approved by the FDA if they have not already gained
status as GRAS (Generally Regarded as Safe).le If a food additive is not
proven safe by the entity seeking to introduce it into the food suppl¡
beverages containing such additives are considered 'adulterated' and
may be condemned by the FDA.20 Conversel¡ manufacturers of dietary
supplements are responsible for determining their products' safety
without any DSHEA requirement to obtain pre-approval for an ingredient unless it is new. Thus, ingredients not designated GRAS are found in
some energy drinks labeled as dietary supplements.
Owing to these labeling issues, it is difficult to determine amounts
of many ingredients contained in energy drinks. Table r summarizes
calorie, sugar, caffeine, and sodium content of prominent, nationally
advertised iog"r-.*..t ned energy drinks identified in a ¿oro study.4 On
the basis of the labels of these products, the most common additional
ingredients are sodium compounds, guarana, panaxginseng, and taurine.
Sugar and sugar substitutes
A comprehensive study of energy beverages reported that the median
sugar content of sugar-sweetened energy drinks was z7 g per 8 oz.
serving, comparable to sodas and fruit drinks, and higher than sports
drinks and flavored water.a\üüith one exception, all energy drinks in this
analysis were available in large, non-resealable containers, providing
excessive sugar and calories in a single serving. Sixty-nine per cent of
energy products also contained artificial sweeteners in lieu of or in
addition to sug"r.a More than half of these were not labeled as diet
products; digt labels would normally alert consumers to the presence of
artificial sweeteners.
Consumption of sugary beverages is associated with increased risk for
dental caries, weight gain, overweight, obesit¡ diabetes, and heart
disease.2l In zoo8, sugary beverages made up j r per cent of added sugar
in the diet of 6-rr year olds and 44per cent ofthe added sugar consumed
by rz-r7 year olds in the United States.22 Although added sugar intake
derived from sugary beverages in total, such as soda, has decreased since
ry99, addedsugar intake from energy drinks has increased.22 Consistent
with sales data, youth may be substituting energy drinks for other sugary
beverages.2'3
'!ì zor3 Macmillan
Publishers Ltd. ory7-5897 Journal of Public Health Policy YoL 34, z,
L54-z7a zi7
P
p
o
j
Þ
Table
r:
Caffeine, calorie, sugar, and sodium content of common sugar-sweerened energy drinks"
Productb
Additional Manufacturer
-*
\þ
\
o
Þ
È
Þ
o
t
F
N
{
I
Þ
Amp Energy
AZEnergy
Full Throttle (Red Berry)
Monster Energy
Monster Energy
Monster Energy
NOS
Red Bull
Red Bull
Red Bull
Red Bull
Rocksta¡
Rockstar
Rockstar
Venom Energy (Black Mamba)
4
PepsiCo
t
Arizona
24
z4
z4
4
o
o
o
II
IT
II
t
ABA member Cdn size
conlþdny
uarieties'
Coca-Cola
Hansen Beverage Company
Coca-Cola
Red Bull
Red Bull
Red Bull
Red Bull
Rockstar
Rockstar
Rockstar
Dr. Pepper Snapple
X
(oz.)
t6
_
X
X
X
X
X
Sugarper Sodiumper
cdrt
(g)
can (mg)
2ZO
58
¡88
zo
t6
t6
49
zoo
230
200
58
!4
54
t6o
r8o
240
8¡
270
J-
320
z6o
300
400
t6
8.4
T2
r6
20
:
Cdlories þet
cøn (kcal)
f42
¡88
r5
X
per
(nzg)
Caffeine
can
I
t6
t4
t6'9
¡6o
aro
r40
ro8
360
54
27
4ro
f4a
r89
237
4o
8o
rro
II4
r54
r6o
L9z
8o
t75
39
54
68
f40
3L
t6o
z8o
6z
8o
/40
4zo
93
IZO
L70
2.50
57
3zo
99
nNutrition information as of September
eorz for each available cansize for nationally adverrised energy drink brands identified in the zorr Sugary
Drink FACTS report from the Rudd Center for Food Policy Ec Obesiry,
olnformation given for
original variery of drink brand, For those brands tlrat do nor have an original variet¡ the flavor is specified.
"Number includes additional sugar-sweetered unique flavor varieties withín each listed brand, not including multiple can sizes,
Regulating energy
clrinks
*
Caffeine
Energy drinks are touted for high caffeine content, but manufacturers do
not always report the amount in each container. In the zoro study of
sugary drinks, 54 per cent of 83 total energy drink products reported
their caffeine content with a median of 8o mg per 8 oz. serving or shot,
more than double the median caffeine in 8 oz. of soda.a Two products
contained extreme levels and were available in zo oz. containers,
providing 245 mg and 325 mg of caffeine.a Another study found that
energy drinks may contain up to 5o5 mg of caffeine per container.e
Caffeine toxicity is a concern for youth.In zoo7, there were 5448
caffeine overdoses reported in the United States and a strikin g 46 per cent
of them occurred in persons younger than 19 years.s The AAP raised
additional concerns for children because of caffeine's effect on developing neurological and cardiovascular systems, plus a risk of physical
dependence and addicti on.r2 Caffeine binds to cell membranes in place
of adenosine, an inhibitory neurotransmitter, causing changes in normál
physiological processes. Specific effects of caffeine consumption include
disturbed sleep, increased body temperature and gastric secretions,
increased blood pressure and heart rate, as well as a risk of physical
dependence and addiction. This is especially problematic for youth
because they are still growing. The AAP specifically cautioned that
dietary intake of caffeine can produce harmful adverse effects in youth
and should be'discouraged for all children'.12
Sodium and other ingredients
of sodium. In the zoro
stud¡ the median sodium level was rz3mg per 8 oz. serving or shot,
more than three times the amount in soda.a Several energy drinks had
Energy drinks contain surprisingly high levels
even more extreme levels,
with one containing 340mgper
8 oz. serving.a
Diets high in sodium can result in high blood pressure and increased risk
for heart disease and stroke.23
Energy drinks often contain specialry ingredients with purported health
benefits, but that can have negative effects on young people. Table z
provides information on drree of the most conunon ingredients: guarana,
taurine, and panax ginseng. Many of the same novelty ingredients found
in energy drinks are also ingredients in over-the-counter diet drugs.27
As consumption of energy drinks increases, these ingredients raise
.Q
ror3 lvlacmillan
Publishers
Ltd. oa97-5897 Joumal ofPublic Health Policy Vol. 34, z,
zs4-z7r
zS9
Poneranzetal
X
Table
z: Conmon
lngrediert
energy drink ingredients
lntended
effcctss Generally
the
Comments front
recognized as American Acndenty
safe (GRAS) of Pediatrics clinical
Other notes
reþorfa
Guaran¡
Stimulant (caffeinecontaining)
Guarana is conceming Conrains 4o milligrams
for youth because it
of caffeine p€r €iranr
increases the
Taurine
Arnino acid believed
to assist rvith cell
metabolism,
thought to
improve rrthletic
performance
Pana-r ginseng Thought to irnprove
athletic
performance
tot¡l
amount of caffeine
in the product
Amino acids in energy
drìnks should be
No
Mayo Clinic study found
discouragecl in
no evidence thar ir
produces trdvenised
children
benefitl't
Not Availeble
Potential negrtive side
effects include
insomnia, menstruirl
problems, increased
heart rarc, and blood
pressure disru rbances26
significant concerns because it is unclear what combined health impact
they may have on consumers, especially youth.
Marketing
A comprehensive analysis of marketing practices and youth exposure to
this marketing in the United States confirmed that several energy drink
manufacturers market their products using media and techniques aimed at
adolescents.4In zoro, US adolescents saw on âverage rz4 tãlevision ads
for energy drinks and shots, which is the equivalent of one ad every 3
days.a This is similar to adolescents'viewing of regular soda ads (rzz), ãnã
more ads for energy drinks and shots than seen by adults.a Adolescents
viewed 9-16 per cent more ads than adults for three energy drink brands.28
The majority of energy drink ads viewed by adolescenrs appeared on
youth-targeted cable networks including Adult Swim (8o-9o per cent
more adolescent thân adult viewers), MTV and MTVz (88-199 per cent
more adolescent viewers), and Comedy Central (zo-3o per cent more
adolescent viewers).28
z6o
íQ zotS Macmillan Publìshers Ltd. ot97-5897 Joumal of Public Health Policy Yol.
34 z,254-z7t
*
Energy drink brands also sponsor extreme sports competitions and are
prominent in digital media that disproportionately appeals to adolescents. Adolescents were approximately twice as likely to visit the
Monster and Rockstar energy drink websites compared to adults 14 and
youth under age 18 often visited Facebook pages of popular energy
drinks, comprising r r per cent of unique visitors for Red Bull and 3 8 per
cent to Monster's page.2e Although it does not appeâr that energy drink
companies directly market to children less than r2 years of age, many
children view the same media as adolescents. As a result, children in the
United States saw on ayerage 6z energy drink and shot ads in zoro,
which is on par with the number of ads they saw for the children's drinks
Capri Sun and Kool-Aid.a
Support for Regulation
In zoo8, scientists and physicians wrote to the FDA requesting increased
regulation of energy drinks because their high caffeine content puts
youth at risk for caffeine intoxication and alcohol-related injuries.sO
France, Denmark, and Norway attempted to ban Red Bull because of
concerns about excessive caffeine and other novel ingredients in the
product,3l but the European Court of Justice found it to be an improper
trade restriction.32
In zorr, Canada officially designated energy drinks as subject to
regulation as food; they established specific criteria, including composition restrictions and labeling requirements.33 Canada determined the
maximum amount of caffeine permitted per single-serve container to be
r8omg and designated all non-resealable containers one seruing.33
Canada also requires labels to disclose the amount of caffeine per serving
and to include warnings for use by children and certain sensitive adults.r3
The Rudd Center for Food Policy & Obesity conducted a nationally
representative online survey of 98 5 US parents of z-r7 year olds in zor r,
seeking to understand attitudes about energy drinks, beliefs about
appropriateness of these drinks for their children, feelings regarding
caffeine and other common ingredients, and attitudes toward energy
drink labeling and regulation.3a They found rhat 67 per cent of parents
were concerned about the caffeine content of beverages for their
children, 78 per cent agreed that energy drinks should not be marketed
to children and adolescents, and 74 per cent agreed these drinks should
not be sold to children or adolescents. In addition, I5 per cent of parents
r.Qror3N{acmillanPublishersLuJ.otgT-5897
JoumalofPublicHealthPolicyYol.34,z,L54-2Tr 26l
*
Ponteranzetal
agreed that regulations requiring reporting
labels were warrânted for energy drinks.
of caffeine and warning
In zotz, US Senators Durbin and Blumenthal asked the FDA
for increased regulation of energy drinks, including clarifying labeling
requirements, directly regulating the amount of caffeine permitted in
products, and an FDA determination of the safety of other additives and
ingredients.35
Regulatory Recommendations
The FDA has primary authoriry over the safet¡ labeling, and ingredients
of energy drinks.36 Federal làw preempts state and lõcal governments
from addressing issues in the FDAt domain. state and local governments
(collectively states), via their legislatures and agencies, can, however,
exercise authority over public health and safety to regulate the sale of
these products and protect consumers.3T lf a government entiry determines that increased regulation of energy drinks is warranted, several
options are available, summarized in Table 3 and discussed below.
Designation as beverages
The FDA issued a non-binding draft guidance document in zoog distinguishing beverages from liquid dietary supplements,l6 and the agency
is currently finalizing the guidance documenr.3t Th. FDA has explãined
that even if a manufacturer characterizes a product as a dietary
supplement, it may be a beverage for regulatory purposes. Beverages
can be distinguished by packaging, volume, advertising, name, and
similarity to other beverages (for example, soda),16 whereas a dietary
supplement is defined as 'a product taken by mouth that contains a
"dietary ingredient" intended to supplement the diet'.16 According to
the FDA, energy drinks labeled as supplemenrs are mislabeled.
Ingredients
The FDA expressed concern that energy drinks contain some GRAS
ingredients 'at levels in excess of their traditional use levels', which
'raises questions regarding whether these higher levels and other new
conditions of use are safe'.76 The FDA granted GRAS sratus to added
srgar3s and caffèine (at levels of o.oz p.r.rrrt of the product) in the
z6z
(9 zory lvfacmillan Publishers Ltd. o197-5897 Joumal of Public Health poticy Vol.
3-1,
z, zr4_z7t
Regulating energy
Table
3: Potential interventions to
'lopic
reduce underage consumption
a
o
Labeling o
.
.
o
o
o
o
Retail
*
ofliquid energy products
Actor
lnteruention
Ingredients o
drinks
Reconsider GRAS status for problernatic ingredients
(including caffeine, sugar, and guarana), especially in
large quantities
Add limirations to permissible amounts of GRAS
ingredients
Thke enforcemenf action against manufacturers that
add unapproved ingredients
Require caffeine disclosures on all products regulated
by FDÄ.
Establish Daily Reference Value (DRVs) f'or caffeine and
added sugar
Require rvarning labels for liquid energy proclucts
Require liquid energy products cornply with the NLEA
Take enforcement actions against products r¡islabeled
as dietary supplelnents
Take enforcement acrion against the marketing of
mislabelecl products or products with false or deceptive
FDA
FDA
FDA, ACs
FDA
FDA
FDA
FDA
FDA, AGs
FTC, AGs
o
claims
Rcquire age lirnits for
r
o
o
Establish location restrictior-rs in retail establishmer-rts State, Local
Prohibit the sale oithe most problematic products
State, Local, AGs
[.stablish excise taxes on highly sugared products
Congress, State,
purchase
Congress, State,
Local
Local (to extent
authorized)
Marketing a
Resc¿rch o
o
Stop marketing to adolescents, including on
programming and in events that appeal to then
Mcasure population caffeine consurnption and youth
corÌsumption of energy drinks and shots
ldentify best practices to reduce sales to underage
corìsurners
ABA, Manufactr.rrers
Public Health
Community
Policy Advocates
rg7os.3e During the approval process, the Select Committee on GRAS
substânces recognized potential health hazards associâted with consuming added sugar at levels higher than ât that time and caffeine in doses
larger than used in cola-type beverages.3s'4O E tergy drinks contribute to
high added sr¿gdr conslrmption, which exceeds the levels at the time of
GRAS approvâI, and they contain far more caffeine than cola-type
beverages.2z Further, although the stimulant guarana is GRAS up to a
specified amount, it is unclear exactly how much guarana is in energy
drinks and how much would be considered safe when it is added to an
already highly caffeinated product.
,.(;
:or3 lr{acmillan
Publishers
Ltrl. or97-5897 Joumal of Public Health Policy Vol. 34, z, 254-z7t
261
The FDA has the authority ro revise GRAS status for sugar, caffeine,
and guarana and to regulate the amount of each ingredient permitted to
be added to beverages. The agency can mandate maximum levels of these
ingredients in single-serving containers.
The FDA also expressed concern that other ingredients in energy
drinks are nor GRAS.and are nor being used in accord with existing foðd
additive regulations.r6 Taurine and panax ginseng, among other potential ingredients, are nor approved for use in beverages. The FDA has the
authority to designate these products as adulterated and unsafe for
the food supply.16 The agency cãn reprimand manufacturers or condemn
the products outright.
Labeling
The us government has several labeling options that should be considered to protect and inform consumers abour the ingredients and risks
associated with energy drinks. congress can amend the FDCA and the
FDA can issue binding regulations thar energy drinks musr be labeled as
beverages and that caffeine content must be disclosed on all products
under the FDA's purview.4l
some or all energy drinks should contain warnings about caffeine
toxicity and the introduction of ingredients not normally found in the
food supply. Toda¡ when caffeine is added to srimulant drug producrs,
the package musr bear a specific warning label stating that thã produc is
for 'occasional use only' and not intended for child¡en under rz years of
?gr.:' US law requires a warning when 'foreseeable risks of harm posed
by the product could have been reduced or avoided by the provision of
reasonable instructions or warnings' and the omission of such a warning
'renders the product not reasonably safe'.a3 ER data from visits involving energy drinks, show these products may be regarded as not reasonably safe without warnings.
Consumer protection actions
(Frc) and state arrorneys general (AGs)
consumer protection actions to address
labeling and ingredient violations identified above. The FTC can bring an
action against manufacfurers for unfair and deceptive marketing practices.
The state AGs have similar authoriry over quesrionable marketing and
The Federal rrade commission
have authority
to institute
264 iþ zot3 Macmill¿n
Pnblishers
Ltd. o197-5897 Journal of public Health policy vol. 14, z. zt4-zla
Regulating energy
clrinks
*
labeling and can additionally bring actions to protect citizens from
particularly problematic products.aa In zorz, for example, New York's
Attorney General started an investigation into whether energy drink
manufacturers were misleading consumers about caffeine content and
potential health risks.as
Retail restrictions
State governments
in the United
States may enact retail regulations.
Seventy-nine per cent of energy drinks are sold from convenience stores,
and thus subject to a variety of potential regulations.a States can, for
example, restrict the sale of energy drinks to youth under a certain age; an
option supported by parents. In zoro, a New York county legislator
proposed a ban on the sale of energy drinks to minors younger than 19
yeats.46 Lawmakers can determine which age is appropriate. Implementa-
tion would be straightforward, because retail outlets are akeady legally
required to verify the age of customers purchasing alcohol and tobacco.
Another option would be to regulate the location of problernatic
products in the retail environment, akin to state requirements that
robacco be sold from behind the counter. Energy drinks are generally
offered in a refrigerator case near alcoholic or other sugary beverages.
This placement may imply that they are similar to sugary beverages and/
or encourage consumers to mix them with alcohol. Research might help
determine how revised placement of drinks could have a positive impact
on public health by discouraging purchases and the mixing with alcohol.
Research can answer the question whether the top shelf of coolers or
aisles, the back of the store, or behind the counter would help protect
consumers.2l
Another retail restriction would ban the sale of certain energy drinks,
such as those in large non-resealable containers or with the highest
caffeine content. A bill proposed in Oregon sought to ban sale of 'highcalorie' beverages in single-serving containers larger than rz oz.a7 The
same rype of restriction could be placed on the sale of highly caffeinated
products in large containers.
Finall¡ it is noteworthy that an excise tax placed on sugary beverages
would surely âpply to sugâry energy drinks. The underlying rationale
and potential benefits of such a tax have been discussed elsewhere; the
goal is to decrease consumption.l Both federal and state governments
can institute excise taxes. Local jurisdictions cân sometimes also enact
,d-r
¿o¡3 À{acmill¡n Publishers Ltd. org7-5897 Joumal of Public Health Policy Yol.
y, z,254-z7r
265
*
Po¡teranzetal
taxes or fees
localities.2l
- to the extent permitted
by the stâte's laws governing
Marketing restrictions
Tighter regulations on the marketing of energy drinks to adolescents are
warranted, but in the United States a substantial barrier exists to government enâcting such regulations. The Supreme Court has interpreted the
First Amendment of the Constitution to protect marketing, or comntercial speech, from government interference. Thus, the United States
has focused on self-regulation, hoping to maintain some conffol over
marketing directed ar yourh.
The ABA established guidelines for the sale and marketing of energy
drinks, under which member companies agree to refrain from marketing
producls to children (ages z-r r) and selling them in schools (grade levels
K-rz).18 The guidelines also srate that ãn..gy drinks shoold nor be
promoted âs sports drinks or in connection with alcohol consumption.
In response to criticism of marketing thât promotes energy drinks to
youth, both Red Bullas and the ABA,4e as a spokes-organiiation for its
member companies, reiterated that they do not market energy drinks to
children under age rz. But these self-regulatory pledges do nor prohibit
marketing targeted directly to adolescents and, as noted, despite these
restrictions, children and adolescents continue to be exposed to large
numbers of advertisements for energy drinks.
Self-regulation of alcohol marketing ro minors (zo years and younger)
provides a potentiâl blueprint for reducing energy drink marketing ro
youth. The FTC has recommended a self-regularory approâch to reduce
underage exposure to alcohol marketing. Major alcohol suppliers agreed
that they would not advertise in media with an audience comprising
more than 3o per cenr minors and have largely complied.so The National
Research Council (NRC), Institute of Medicine (IOM),51 and 19 state
AGss2 recommended tighter self-regulatory standarás, incl.rdír,g rro
alcohol advertising in media with an underage audience share of r j per
cent (approximately their share of the us population) and restricrions on
marketing practices with substantial underage appeal. The NRC and
IOM also recommended establishment of an independent review board
to monitor alcohol marketing practices. A similar protocol would work
well for energy drinks.
266
tÇ zot3 Macmillan Publishers Ltd. or97-t897 Joumal of Public Health policy Vol. 34, z,
zr4-z7t
Rcgulating energy
rlrinks *
Companies that belong to the ABA currently comply with their selfregulatory commitments, but this program has limitations. Several of the
highest selling energy drink brands do not belong to the ABA. At a
minimum, these companies should agree to abide by ABA guidelines.
However, to address the majority of youth-targeted marketing of energy
drinks, all energy drink manufacturers should also agree to discontinue
their marketing practices that disproportionately appeal to adolescents,
including advertising on television programming with a higher-thanaverage proportion of youth in the audience and the use of social media
and sponsored events.
Discussion and Conclusion
Existing evidence points to significant public health issues arising from
youth consumption of energy drinks, but further research and analysis
are needed:
More comprehensive measurement of youth consumption of caffeine
and energy drinks, separate from other sugary beverages. Because
energy drinks are relatively new products in the American marketplace, ongoing dietary measurement panels do not adequately monitor and report on these products.
Research to determine consumer understanding of ingredients and
claims on energy drink labels would help us understand the extent to
which current practices mislead or deceive.
Studies of energy shots are also warranted. We know little about
energy shot consumption by youth; but 8z per cent of the energy
product ads viewed by children and adolescents promoted one
shot: 5-Hour Energy.a Of all products examined in the zoro stud¡ a
z.s oz. shot had the highest per-serving caffeine content overall,
zoo mg.a Manufacturers designate energy shots as dietary supplements so they are located with other díetary supplements in pharmacies, which may send an unwarranted health message to consumers.
In other retail outlets, shots are often located in free-standing displays
at the check-outa further encouraging purchase. The FDA should pay
particular attention to categorization and labeling of shots because
companies market them in media viewed by youth and they contain
extreme levels of caffeine that could be dangerous for children and
adolescents.
ç
aor3 lr{acmillan Publishers Ltd. o197-5897 Jomal of Public Health Policy Yol. 34,
z,254-z7r
267
o To identify best policies, research might help local jurisdictions
determine the best location in retail establishments ro require problematic products to be placed to discourage purchase by youth.
Alternativel¡ locales can experiment with product placement restrictions to determine which locations work best.
consumption of energy drinks * , nrot" health concern especially for
young people. Increased regulation is warranted to inform and protect
consumers by addressing problematic ingredients, clarifying labeling
requirements, and restricting youth âccess. At a minimum, increased selfregulatory efforts should be instituted to protecr youth from marketing.
Energy drinks are a unique beverage and should be regulated accordingly.
About the Authors
Jennifer L. Pomeranz (JD, MPH) is the Director of Legal Initiarives at
the Yale Rudd center for Food Policy & obesiry ar yale university.
Ms. Pomeranz speaks and publishes on subjects inclucling: sugar
labeling, weight discrimination, food marketing to children, the First
Amendment, preemption, regulating sugary beverages, and innovative
legal solutions ro obesity. she earned her Juris Doctorare from cornell
Law school, and her Master of Public Health from the Harvard school
of Public Health.
christina R. Munsell (MS) is a Research Associate and Registered
Dietitian at the Rudd center for Food Policy Ec obesity at yale universiry.
she completed a dietetic internship and Masrer's degree in Health care
Policy and Management at Stony Brook University after earning a
Bachelor's degree in Nutrition from Texas christian university. Munsell
works with the marketing ream ar the Rudd cenrer, develàping and
executing projects that analyze food marketing to youth as well as
providing nutrition expertise for other Rudd Center projects and research.
Jennifer L. Harris (MBA, PhD) is Director of Marketing Initiatives
at the Rudd center for Food Policy & obesiry at Yale university. she is
responsible for identifying and coordinating research initiatives to understand and communicate the extent and impact of children's exposure
to food advertising. Dr Harris received her BA in Political sciencã from
268 (Qzotl MacmillanPublishersLt<I. ot97-5897 JournalofPubticHealthpolicy yol.34z,254-z7t
Regulating energy
clrinks *
Northwestern Universiry her MBA in Marketing from The Wharton
School at the University of Pennsylvania, and her PhD in Socíal Psychology
from Yale University. Before returning to graduate school, she worked for
18 years as a Vice President in marketing at American Express and ran a
marketing consulting firm specializing in marketing strategy and new
product and market development. Dr Harris has written on the psychological and behavioral effects of advertising to children and adolescents
and conducted research to quantify the amount and types of food
marketing seen by young people and its impact on their health and diet.
References and Notes
r.
Brownell, K.D. et al (zoo9) The public health economic benefits of taxing sugar-sweetened
beverages. New Englaul Journal of Medicine 36r(úl: r599-t6o5.
z. PRNervsll'ire.(zorr)Energydrinkindustryreportshowsstronggrowth.zoDecernber,http://
w
rvrv.prnews\\'i re.com/news-releases/energy-drink-industry-report-shorvs-strong-grolvth-
r359r5z63.html, accessed z6 September zorz.
3. BEVNET. (z.orr) Morgan Stanley report: Energy drinks and RTD teas perform rvell in yearover-yeâr comparison. 8 Decernber, http://www.bevnet.cour/news/zor r/morgan-stanley-reportaccessed z6 September
energy-drinks-and-nd-teas-pertbrm-rvell-in-1's¿¡-over-year-comparison,
20L2,
.1. )'ale Rudd Cenrer for Food Policy & Obesitv. (zor r) Sugary drink FACTS: Evaluating sugaty
drink r¡utrition and marketing to youth, http://wrvw.sugarydrinkfacts.org/resourceVSugaryDrinkFACTS-Report.pdf, accessed z6 September zorz.
5. lltnsen Beu. Co. v. lnnou¿ttion Ventures, LLC, zoog U.S. Dist. LEXIS rz76o5 (S.D' Cal.
:r
December, zoog).
I-{.À,t., Sherr¡ 8., Brener, N. and O''Ibole,'f. (zorz. ) Factors âssociated with
sugar-sn'eerened beverage intake among united states high school students. Joru'nal ofNutrition
6. Park, S., Bl;rnck,
14z(r);3o6-3r2.
7. Sinron, M. and Mosher, .f. (zoo7) Alcohol, Energy l)rinks, and Youtb: A Dangerous Mix.
Crlifornia: M¡rin Institute.
8. Seifen, S.lt4., S,chaerchter, J.L., Hershorin, E.R. and Lipshultz, S.E. (zorr) Health
efftcrs of energy drinks on children, adolescents, and young adults. Pediatrics tz7(31:
5
r r-5 28.
9. Zenirb ftrternation¿I. (zoo7) US overtakes Thailand as 'rvorld leader in energy drinks, http://
ro.
tt.
rz.
r3.
www.zenithinternational.cor¡r/news/press-release-detail.asp?id=2o6, accessed z6 September
20L2.
Reissig, C..f., Strain, E.C. and Griffiths, R.R. (zoo9) Caffeinated energy drinks - A growing
problenr. Drugand Alcohol Dependence 99(r-3): r-ro.
Hdnsen Beu, C<;. v. Innouation Ventures, LLC, zooS U.S. Dist. LEXIS 762ç (S'D. Cal'
z8 September, zoo8).
(zorr) Sporrs drinks and energy drirrks for children and adolescents: Are they appropriate?
Pediatrics 1" 7(61: r r8z- r r89.
Substance Abuse and Mental l-Iealth Services Adrninistration, Center for Behavìoral Health
Statistics and Quality. (zorr) The DAVN Report: Emergency Department Visits Involving
Energy Drinks. zz November, R<rckville, MD.
C ¡or-t lv{acmillen
Publishers
Ltd. oa97-5897 Joumal of Public Health Policy Vol. 34, z, L54-L7r
269
ponTerãnzetrl
*
r4'
WA State Liquor Control Board. (zoro) The mix with dangerous risks: Energy dri¡ks a¡l{
alcohol, http://liq.rva.gov/education/alcohol-energy-drinks,
accessed z6 september
zorz.
r5. Gunia,N.arrcl Brorvn,J.A.(zorz) Energydrinks:Healthrisksandtoxicit¡,. MetlicalJournalof
Austnilia ry6(r): 46-49.
16. FDA. (zoo9) Draft guidance for industry: Factors that distinguish liquid dietary supplements
frorn beverages' consideratior.rs regarding novel ingredients, and labeling for beverages and
r7.
other conventional foods, http://s,wra'.fcla.govÆood/GuidanceComplianceRegulator',I;fornìation/GuidanceDocuments/DietarySupplementVucmrg69o3.htrn, accessed z February zor3.
zr U.S.C. ¡+¡(qXs)(F).
r8. ABA.Guidanceiorthe responsiblelabelingandmarketingofenergydrinks.Americ¿nBeverage
Association, http://rvlvlv.ameribev.org/files/319-Enetgyø/ozoDrink%zoGuiclelinesTozoyozS
finalo/"z9.pdt, accessed z February zor3.
r9. zr U.S.C. 3zr(s).
zo. Urrited States v. An Article of Food, Coco Rico, 75zF.zd rr (rst Cir. r9g5).
zt.
Pometanz, J'L. (zotr) Advanced policy clptions to regulate sugar-sr,r'eetened beverages to
support public health. Journal of Public Ilealtb Policy 31j): 7 5-BB.
zz. Welsh,J.A.,Sharma,A.L.G.andVos,M.B.(zorr)Consumptionofaddedsugarsisdecreasing
in tlre United Srates. Anrcrican Jurrnal oi Clinical Nrtritiott 94þ):726-734.
CDC. (zorr)Usual sodiumintakescomparedwithcurrentdietaryguidelines-UnitedStates,
zoo5-zoo8. Morbidity and Mortality Weelzly, zr October, 6o(4r): t4r¡-t417.
24. Cornmittee on Nutrition and the Council on Sports Medicine and Fitncss of the American
4.
Academy ofPediatrics. (zor r ) Sports drinks and energy drinks for childre¡r ancl rdolescents: Are
tlrey appropriat e? Pe d ia tri cs t z7 (61 t ß z- t rB9.
z' 5. Higgins'J.P', Iirttle, T.D, and Higgins. C.L. (zoro) Energy beverages: Conte¡rr
ancl safety'. ùlrr¡'9
Clinic ProceedirzS-s 8 j(rr): rol j-r04Ì.
26. MedlinePlus. (zorr) Panax Ginseng. US National Library of À4edicine, rr Augrrst, http://
rvwrv.nlm.nih.gov/rnedlineplus/druginfo/naturaV¡ ooo.html.
27. Zoller Labs,, LLC v. NBTY ,lnc., rrr Fed. Appx.978 (roth Cir. zoo4).
28. Yale Rucld Center for Food Policy & Obesity. (:.or z) Adolescent-targered televisiçn advc.rtisi¡g
for energy drinks. Februa¡¡ wrvw.yaleruddcenter.org/resourceVuplolcl/doc-s/what/advertisi¡$
TVaclvertisiug_EnergyDrinks_zoro.pdf, accessed z6 September zor z.
29. comScore'SiteDetailReport.Averageofmonthlyuniquevisitorsfrc¡rnOctobertoDecenrber,zorr.
3o. \ffeise,E'(zorz) PetitioncallsforFDAtoregulateenergydrinks.
3r.
US{Totilt¡,zzOctober,http://
www.usatoday.com/newJhealth/zoo8-ro-zr-energy-drinks-N.htrn, accessed z6 September ror z.
Medical Nervs Today. (zoo4) French ban on Red Bull (drink) upheld by European courr.
http://www.meclical'ewstoday.com/releases/5753.php, accessed z6 Seprember
l"f:l*"rL
3z'
33'
D. (zoo8) France reluctantly lifts ban on Red Bull. Australian Food Nervs. r7 July',
lrttp://n'wrv.ausfoodnervs.corn. aul zooSloTl rylfrance-reluctantly-lifts-ban-on-red-bull.html,
accessed z6 September zorz.
Health Canada. (zorr) Flealth Canada's proposed approach to managing caffeinated energy
drinks, http://www.hc-sc.gc.ca./fn-anlegislatiou/poVenergy-drinks-boissons-energisantes-eng.pÈ,
accessed z6 September zorz,
Palmer,
& Obesity. (zorz) Parents' attirudes about energy drinks.
June, http://wwrv.yaleruddcenter.org/resources/uploacVdocs/rvhat/policy/SSBtaxeVSSB_P:r¡enr
_Attitudes_Energy_Drinks.pdf, accessed z6 September zor z.
3 5. Letter to the FDA lìom Senators Durbin and Blumenthal. rr Septemlxr zor z, httpJ/durbin.senate
.gory'public/index.cfm/files/serve?File-id=fe44b7Be-3zae-4rao-Ba6r-zddfr4ab95dr,
acces:sed
34. Yale Rudd Center for Food Policy
z6 Septemlxr zorz.
z7o
tt) zot3 lvlacmillan P¡blishers Ltd. or97-5897 Joumal of Public Health poticy vol.
i4, z, zi4-z7a
Regulating energy
drinks *
t
36. zr U.S.C. $3or et
:
seq.
1,7. Hukbinson lce Cream Co. v. lowa, z4z U.S. r¡3, þ9t6).
38. zr C.F.R. $r84.-r854 et seq.
39. zr C.F.R. $r8z.rr8o.
4o. FDA. (1978) Database of Select Comminee on GRAS Substances Reviews: Caffeine. Repon
No. 89. l.D. code 58-o8-2,
http//www.accessdata.fda.gov/scripts/fcn/fcnDetailNavigation
.cfm?rpt=scogslistìng&id=42, accessed z6 September zorz
4r.
6z Federal Register 49826
(4
September 1997).
42. zr C.F.R. $3+o.jo.
Àmerican Law lnstitute. (1998) Restatement 3d of Torts: Products Liability $z Categories of
Product Defect.
44. Connecticut Á,rtorney General's Office. (zoo7) Attorney general, DCP Cornmissione¡ announce
agreement banishing 'cocaine' drink f¡om CT stores. 7 May, http:llwwwct.gov/AG/cwp/
view.asp?A=27 88&Q= j, 795o4.
45. Schwartz, N.D. (zorz, ) New inquiry into energy drink fitms.The New YorÞTimes, z9 August,
46. Epstein, RJ. (zoro) Under 19? Suffolk bill could ban Red Bull. Newsday. 8 Decembeg http://
4j.
www.newsday.com/long-island/suffolk/under-r9-suffolk-bill-could-ban-red-bull-r.2j:.7343.
47. ORÍlß3zzz(zorol.
48. Montague-Jones, G. (zorr) Red Bull denies child marketing claims in new study. BeverageDaily.com. 3 June, hnp://www.beveragedaily.com/content/view/print/3789r6,
September
accessed z6
zorz.
a9. ABA.(zorr) BeverageindustryrespondstolatestRuddReport,http://www.ameribev.orglfilesl
news/253-ÂBÀ"/"zorespondso/o zotoY"Rudd%ozoReport.pdt accessed z6 September zorz.
5o. FTC. (zoo8) Self-regulation in the alcohol industry: Report of the Federal Trade Commission,
http://wwrv.ftc. govloslzooSloíloSo6z6alcoholreport.pdf, accessed z6 September zo r z.
Research Council and Institute of Medicine. (zoo3) Reducing Underage DrinÞing:
ACollectiue Resporcibilþ. ìüashington DC: National Academies Press.
52. Rowe, G.S. et al (zoo6) A communication from the Chief Legal Oftjcers of the following states:
AZ,CI, DE, HI, ID, IL, IA, ME, MD, NJ, NM, NY, OH, OR, RI, UT, VT, \üü,{, WY. RE:
Àlcohol Reports: Paperwork conÌment FTC File No. Po645o5, 8 Ma¡ http://www.ftc.gov/os/
comments/alcoholmanufacadstudy/5zz85z-orz87.pdf, accessed z6 September zorz.
5r. National
,gì
ror3 Macmillan Publishers Ltd. or97-5897 Joumal of Public Health Policy Yol.34, z, Lj4-z7r L7t
!
ARTICLE
International Food Risk Analysis Journal
Energy Drinks: An Assessment
of the Potential Health Risks
in the Canadian Context
Regular Paper
Joel Rotstein1, Jennifer Barber1, Carl Strowbridge1,
Stephen Hayward1, Rong Huang1 and Samuel Benrejeb Godefroy1,*
1 Food Directorate, Health Products and Food Branch, Health Canada, Canada
* Corresponding author E-mail: samuel.godefroy@hc-sc.gc.ca
ȱ
Received 12 February 2013; Accepted 3 June 2013
DOI: 10.5772/56723
© 2013 Rotstein et al.; licensee InTech. This is an open access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
Abstractȱ Theȱ purposeȱ ofȱ thisȱ documentȱ isȱ toȱ developȱ aȱ
healthȱriskȱassessmentȱonȱenergyȱdrinks,ȱbasedȱonȱhealthȱ
hazardȱ andȱ exposureȱ assessmentsȱ whenȱ consumedȱ asȱ aȱ
foodȱinȱCanada.ȱInȱthisȱdocument,ȱaȱtypicalȱenergyȱdrinkȱ
isȱ exemplifiedȱ byȱ theȱ productȱ knownȱ asȱ Redȱ Bull,ȱ
whereȱ aȱ singleȱ canȱ servingȱ ofȱ 250ȱ mlȱ containsȱ 80ȱ mgȱ ofȱ
caffeine,ȱ1000ȱmgȱofȱtaurine,ȱ600ȱmgȱofȱglucuronolactoneȱ
andȱseveralȱBȱvitamins.ȱȱ
Healthȱ hazardȱ dataȱ onȱ energyȱ drinksȱ wereȱ foundȱ toȱ beȱ
limitedȱandȱthereforeȱtheȱhazardȱassessmentȱwasȱbasedȱonȱ
individualȱ ingredients.ȱ Caffeineȱ wasȱ identifiedȱ asȱ theȱ
ingredientȱ withȱ theȱ greatestȱ potentialȱ forȱ intakesȱ ofȱ
possibleȱ healthȱ concern.ȱ Onȱ thisȱ basis,ȱ excessȱ
consumptionȱ ofȱ energyȱ drinksȱ wouldȱ beȱ expectedȱ toȱ
resultȱinȱhealthȱconsequencesȱsimilarȱtoȱthoseȱfromȱexcessȱ
exposureȱtoȱcaffeine.ȱTheȱmoreȱmildȱandȱtransientȱhealthȱ
consequencesȱ couldȱ includeȱ anxiety,ȱ headacheȱ andȱ
insomniaȱ andȱ theseȱ healthȱ consequencesȱ canȱ becomeȱ
chronicȱ conditions.ȱ Moreȱ severeȱ healthȱ consequencesȱ mayȱ
includeȱ irregularȱ heartbeat,ȱ heartȱ attackȱ and,ȱ veryȱ rarely,ȱ
death.ȱ Currently,ȱ theȱ potentialȱ forȱ taurineȱ andȱ
glucuronolactoneȱtoȱinteractȱwithȱcaffeineȱisȱunknownȱandȱ
thereforeȱ theyȱ mayȱ orȱ mayȱ notȱ exacerbateȱ theȱ effectsȱ ofȱ
www.intechopen.com
caffeine.ȱ Inȱ addition,ȱ theȱ healthȱ effectsȱ ofȱ excessiveȱ intakeȱ
ofȱ taurineȱ andȱ glucuronolactoneȱ areȱ alsoȱ unknown.ȱ Theȱ
healthȱhazardȱassessmentȱconcludedȱthatȱtheȱgeneralȱadultȱ
populationȱ couldȱ safelyȱ consumeȱ 2ȱ servingsȱ ofȱ aȱ typicalȱ
energyȱ drinkȱ perȱ day,ȱ withȱ noȱ healthȱ consequences.ȱ Thisȱ
conclusionȱ wasȱ basedȱ onȱ theȱ safetyȱ ofȱ theȱ nonȬcaffeineȱ
ingredientsȱ ofȱ energyȱ drinksȱ atȱ thisȱ levelȱ ofȱ consumption,ȱ
andȱ theȱ factȱ thatȱ caffeineȱ fromȱ otherȱ dietaryȱ sourcesȱ inȱ
additionȱ toȱ thatȱ inȱ 2ȱ servingsȱ ofȱ energyȱ drinksȱ wouldȱ notȱ
poseȱaȱ healthȱ hazardȱtoȱtheȱ generalȱadultȱ population.ȱTheȱ
consumptionȱofȱenergyȱdrinksȱbyȱsubpopulations,ȱsuchȱasȱ
children,ȱ adolescents,ȱ andȱ pregnantȱ women,ȱ shouldȱ beȱ
limitedȱ toȱ theirȱ respectiveȱ recommendedȱ maximumȱ dailyȱ
intakesȱofȱcaffeine,ȱasȱrecommendedȱbyȱHealthȱCanada.ȱȱ
Usingȱexposureȱmodelling,ȱtheȱpotentialȱhealthȱriskȱposedȱ
byȱ energyȱ drinkȱ consumptionȱ wasȱ examined.ȱ However,ȱ
noȱCanadianȱintakeȱdataȱforȱenergyȱdrinksȱwereȱavailable.ȱ
Therefore,ȱ forȱ theȱ purposeȱ ofȱ modellingȱ intake,ȱ itȱ wasȱ
assumedȱ thatȱ energyȱ drinksȱ areȱ consumedȱ inȱ aȱ mannerȱ
similarȱtoȱthatȱofȱcaffeinatedȱcarbonatedȱsoftȱdrinks.ȱInȱtheȱ
worstȱ caseȱ modellingȱ exposureȱ scenario,ȱ energyȱ drinksȱ
wereȱsubstitutedȱforȱcaffeinatedȱcarbonatedȱsoftȱdrinksȱonȱ
aȱvolumeȱbasis.ȱ
Int.Huang
food risk
j., 2013,
Vol. Godefroy:
3, 4:2013
Joel Rotstein, Jennifer Barber, Carl Strowbridge, Stephen Hayward, Rong
andanal.
Samuel
Benrejeb
Energy Drinks: An Assessment of the Potential Health Risks in the Canadian Context
1
!
Theȱ energyȱ drinkȱ caffeineȱ concentrationȱ wasȱ setȱ toȱ 320ȱ
ppmȱ (80ȱ mgȱ ofȱ caffeine/250ȱ mlȱ serving)ȱ forȱ modellingȱ
purposes.ȱInȱtheȱmostȱconservativeȱscenario,ȱallȱcaffeinatedȱ
carbonatedȱ softȱ drinksȱ wereȱ replacedȱ byȱ aȱ typicalȱ energyȱ
drinkȱ forȱ consumersȱ whoȱ drinkȱ theseȱ beveragesȱ (eatersȱ
only).ȱTheȱresultsȱofȱthisȱconservativeȱestimateȱshowedȱthatȱ
slightlyȱlessȱthanȱ30%ȱofȱmaleȱandȱfemaleȱadults,ȱaboutȱ15%ȱ
ofȱpregnantȱwomanȱandȱmoreȱthanȱ50%ȱofȱtheȱchildrenȱandȱ
adolescents,ȱ amongstȱ consumersȱ whoȱ drankȱ caffeinatedȱ
carbonatedȱ softȱ drinks,ȱ wereȱ aboveȱ Healthȱ Canada’sȱ
recommendedȱmaximumȱdailyȱcaffeineȱintake.ȱȱ
Thisȱ extremeȱ scenarioȱ onlyȱ appliesȱ toȱ thatȱ subsetȱ ofȱ theȱ
populationȱ thatȱ consumesȱ caffeinatedȱ carbonatedȱ softȱ
drinks,ȱwhichȱdoesȱnotȱexceedȱ8%ȱofȱyoungȱchildrenȱ(1Ȭ8ȱ
yearsȱold),ȱ22%ȱofȱolderȱchildrenȱ(9Ȭ14ȱyearsȱold),ȱ32%ȱofȱ
adolescents,ȱ 20%ȱ ofȱ theȱ adultȱ populationȱ andȱ 13%ȱ ofȱ
pregnantȱfemales.ȱ
Inȱ criticallyȱ reviewingȱ theȱ outcomesȱ ofȱ thisȱ exposureȱ
modelling,ȱ itȱ appearedȱ thatȱ theȱ correspondingȱ healthȱ
concernsȱ forȱ childrenȱ andȱ adultsȱ wouldȱ beȱ limitedȱ toȱ
remote,ȱ basedȱ onȱ thisȱ scenario,ȱ inȱ viewȱ ofȱ theȱ parentalȱ
controlȱ thatȱ shouldȱ existȱ andȱ wouldȱ limitȱ accessȱ ofȱ theseȱ
productsȱ toȱ children,ȱ asȱ wellȱ asȱ theȱ abilityȱ ofȱ adultsȱ andȱ
pregnantȱwomenȱtoȱmonitorȱtheirȱownȱcaffeineȱintake.ȱ
Thisȱhypotheticalȱscenarioȱandȱitsȱoutcomesȱcouldȱnotȱbeȱasȱ
easilyȱ excludedȱ forȱ adolescents,ȱ givenȱ thatȱ energyȱ drinksȱ
tendȱtoȱbeȱmarketedȱtoȱthisȱsubsetȱofȱtheȱpopulation,ȱwhichȱ
isȱ lessȱ likelyȱ toȱ adhereȱ toȱ consumptionȱ recommendationsȱ
thanȱadults.ȱTheȱexistenceȱofȱlargerȱvolumeȱcontainersȱ(e.g.,ȱ
710ȱ ml)ȱ increasesȱ theȱ likelihoodȱ ofȱ exceedingȱ caffeineȱ
recommendedȱintakesȱinȱoneȱconsumptionȱsetting.ȱ
Specificȱriskȱmanagementȱmeasuresȱtoȱaddressȱpotentiallyȱ
highȱ caffeineȱ levelsȱ inȱ largerȱ volumeȱ energyȱ drinkȱ
productsȱwouldȱthereforeȱbeȱdesirable.ȱ
Itȱisȱacknowledgedȱthatȱvariousȱdataȱgapsȱwouldȱneedȱtoȱ
beȱaddressedȱtoȱimproveȱtheȱexposureȱassessmentȱandȱtheȱ
overallȱriskȱcharacterization.ȱInȱparticular,ȱdataȱrelatedȱtoȱ
theȱ evolvingȱ consumptionȱ patternsȱ ofȱ theseȱ productsȱ byȱ
theȱvariousȱsubsetsȱofȱtheȱpopulation,ȱinȱCanada,ȱneedsȱtoȱ
beȱgathered.ȱ
Healthȱ Canada’sȱ proposedȱ riskȱ managementȱ approachȱ forȱ
energyȱ drinks,ȱ announcedȱ inȱ Octoberȱ 2011ȱ andȱ updatedȱ inȱ
2012,ȱlimitsȱtheȱconcentrationȱandȱtotalȱamountȱofȱcaffeineȱinȱ
theseȱ productsȱ andȱ requiresȱ thatȱ caffeineȱ andȱ nutritionȱ
informationȱbeȱdisplayedȱonȱproductȱlabels.ȱTheseȱmeasuresȱ
supportȱ theȱ mitigationȱ ofȱ risksȱ relatedȱ toȱ overconsumptionȱ
ofȱ caffeineȱ fromȱ thisȱ typeȱ ofȱ productȱ thatȱ areȱ withinȱ theȱ
possibleȱ areasȱ ofȱ interventionȱ availableȱ toȱ aȱ federalȱ foodȱ
regulator.ȱ Aȱ moreȱ concertedȱ approachȱ (e.g.,ȱ education,ȱ
awarenessȱ andȱ regulation)ȱ andȱ moreȱ researchȱ wouldȱ beȱ
neededȱtoȱascertainȱtheȱeffectivenessȱofȱtheȱvariousȱmeasuresȱ
takenȱbyȱregulatorsȱandȱotherȱstakeholders.ȱ
ȱ
Keywordsȱ
Energyȱ
drink,ȱ
Caffeine,ȱ
Taurine,ȱ
Glucuronolactone,ȱInositol,ȱBȱvitaminȱ
Disclaimer:ȱȱ
ȱ
Thisȱ documentȱ wasȱ developedȱ asȱ aȱ reviewȱ ofȱ theȱ
scientificȱinformationȱavailableȱpertainingȱtoȱtheȱsafetyȱofȱ
ingredientsȱ thatȱ mayȱ beȱ partȱ ofȱ theȱ compositionȱ ofȱ theȱ
beveragesȱ knownȱ asȱ Caffeinatedȱ Energyȱ Drinksȱ (Energyȱ
Drinks).ȱ Theȱ documentȱ wasȱ developedȱ duringȱ aȱ periodȱ
spanningȱfromȱ2010ȱtoȱlateȱ2011.ȱAȱnumberȱofȱreferencesȱ
andȱnewȱstudiesȱhaveȱbeenȱmadeȱavailableȱsinceȱthenȱandȱ
areȱnotȱreferencedȱinȱtheȱpresentȱdocument.ȱAnȱupdateȱofȱ
theȱpresentȱHealthȱRiskȱAssessmentȱisȱenvisagedȱinȱ2014Ȭ
15,ȱ uponȱ reviewȱ ofȱ informationȱ currentlyȱbeingȱcollectedȱ
byȱHealthȱCanada’sȱFoodȱDirectorateȱsinceȱtheȱregulatoryȱ
decisionȱmadeȱinȱOctoberȱ2011,ȱtoȱregulateȱEnergyȱDrinksȱ
asȱ beveragesȱ (i.e.ȱ food)ȱ asȱ opposedȱ toȱ theirȱ formerȱ
classificationȱasȱNaturalȱHealthȱProducts.ȱȱ
ȱ
1.ȱIntroductionȱ
1.1ȱȱPurposeȱ
Theȱpurposeȱofȱthisȱdocumentȱisȱtoȱprovideȱaȱhealthȱriskȱ
assessmentȱ ofȱ energyȱ drinkȱ productsȱ basedȱ onȱ theirȱ
consumptionȱ asȱ beveragesȱ inȱ Canada.ȱ Theȱ documentȱ
attemptsȱ toȱ provideȱ aȱ scientificȱ analysisȱ basedȱ onȱ theȱ
informationȱ availableȱ toȱ date,ȱ whichȱ couldȱ thenȱ beȱ usedȱ
toȱsupportȱtheȱdevelopmentȱofȱsuitableȱriskȱmanagementȱ
measuresȱforȱtheseȱproductsȱwhenȱconsumedȱasȱfoods.ȱȱ
2.ȱDefiningȱaȱtypicalȱenergyȱdrinkȱȱ
2.1ȱDefiningȱproductsȱknownȱasȱenergyȱdrinksȱ
Forȱtheȱpurposeȱofȱthisȱdocument,ȱaȱtypicalȱenergyȱdrinkȱ
isȱ characterisedȱ byȱ theȱ ingredientsȱ andȱ ingredientȱ levelsȱ
shownȱ inȱ Tableȱ 1.ȱ Ingredientsȱ includeȱ caffeine,ȱ taurine,ȱ
glucuronolactone,ȱ inositolȱ andȱ aȱ varietyȱ ofȱ Bȱ vitamins.ȱ
Theȱ basisȱ forȱ thisȱ characterisationȱ isȱ theȱ formulationȱ ofȱ
theȱ mostȱ commonlyȱ soldȱ productsȱ ofȱ thisȱ categoryȱ ofȱ
beverages,ȱ suchȱ asȱ theȱ productȱ knownȱ asȱ Redȱ Bull,ȱ
whichȱwasȱtheȱfirstȱenergyȱdrinkȱapprovedȱforȱtheȱmarketȱ
inȱ Canada.ȱ Basedȱ onȱ 2009ȱ marketȱ dataȱ ofȱ theȱ globalȱ
marketingȱ researchȱ firmȱ ACNielsen,ȱ Redȱ Bullȱ
comprisedȱ approximatelyȱ 37%ȱ ofȱ theȱ salesȱ ofȱ energyȱ
drinksȱ inȱ Canadaȱ and,ȱ asȱ such,ȱ hadȱ theȱ largestȱ marketȱ
share.ȱThisȱisȱconsistentȱwithȱaȱmoreȱrecentȱreportȱ(AAFC,ȱ
2011)ȱ thatȱ statesȱ thatȱ Redȱ Bullȱ constitutesȱ 43%ȱ ofȱ theȱ
salesȱofȱenergyȱdrinksȱinȱNorthȱAmerica.ȱ
ȱ
Althoughȱ Healthȱ Canadaȱ doesȱ notȱ haveȱ aȱ definitionȱ orȱ
standardȱ forȱ productsȱ knownȱ asȱ energyȱ drinks,ȱ otherȱ
foodȱregulatorsȱhaveȱreachedȱaȱdecisionȱonȱthisȱaspectȱofȱ
theseȱproducts.ȱForȱexample,ȱAustraliaȱandȱNewȱZealandȱ
categorizesȱ energyȱ drinksȱ asȱ ‘formulatedȱ caffeinatedȱ
beverages’ȱ whichȱ areȱ definedȱ underȱ theȱ Australiaȱ Newȱ
Zealandȱ Foodȱ Standardsȱ Codeȱ asȱ “aȱ nonȬalcoholicȱ waterȬ
basedȱ flavouredȱ beverageȱ whichȱ containsȱ caffeineȱ andȱ
ȱ
2
Int. food risk anal. j., 2013, Vol. 3, 4:2013
ȱ
www.intechopen.com
!
mayȱ containȱ carbohydrates,ȱ aminoȱ acids,ȱ vitaminsȱ andȱ
otherȱ substances,ȱ includingȱ otherȱ foods,ȱ forȱ theȱ purposeȱ
ofȱenhancingȱmentalȱperformance”.ȱȱ
Ofȱ theȱ energyȱ drinkȱ productsȱ thatȱ wereȱ inȱ theȱ NHPȱ
submissionȱ queue,ȱ approximatelyȱ 13%ȱ fellȱ withinȱ theȱ
‘typicalȱ description’ȱ forȱ energyȱ drinks.ȱ Basedȱ onȱ marketȱ
ȱ
shareȱdataȱfromȱACNielsenȱ(2009),ȱRedȱBullȱwasȱrankedȱ
theȱhighest,ȱcomprisingȱapproximatelyȱ37%ȱofȱtheȱmarket.ȱ
Inȱ Europe,ȱ theȱ Irelandȱ Foodȱ Safetyȱ Promotionȱ Boardȱ
(FSPB)ȱ establishedȱ aȱ committeeȱ consistingȱ ofȱ externalȱ
expertsȱtoȱresearchȱtheȱhealthȱeffectsȱofȱ‘stimulantȱdrinks’ȱ
(energyȱ drinks).ȱ Theȱ committeeȱ notedȱ inȱ itsȱ finalȱ reportȱ
thatȱ thereȱ isȱ noȱ agreedȱ definitionȱ inȱ theȱ regulatoryȱ
frameworkȱ forȱ productsȱ referredȱ toȱ asȱ energyȱ drinksȱ orȱ
‘stimulantȱ drinks’.ȱ Forȱ theȱ purposesȱ ofȱ theȱ report,ȱ theȱ
termȱ ’stimulantȱ drinks’ȱ wasȱ adopted.ȱ Theȱ committeeȱ
definedȱ stimulantȱ drinksȱ asȱ “beverages,ȱ whichȱ typicallyȱ
containȱcaffeine,ȱtaurineȱandȱvitamin(s),ȱandȱmayȱcontainȱ
anȱ energyȱ sourceȱ (e.g.ȱ carbohydrate),ȱ and/orȱ otherȱ
substance(s),ȱ marketedȱ forȱ theȱ specificȱ purposeȱ ofȱ
providingȱ realȱ orȱ perceivedȱ enhancedȱ physiologicalȱ
and/orȱperformanceȱeffects”ȱ(FSPB,ȱ2003).ȱ
2.2ȱȱMajorȱcomponentsȱofȱenergyȱdrinkȱproductsȱȱ
Duringȱ theȱ preparationȱ ofȱ thisȱ manuscript,ȱ energyȱ drinksȱ
wereȱ marketedȱ inȱ Canadaȱ underȱ theȱ Naturalȱ Healthȱ
Productsȱ(NHP)ȱRegulations.ȱAtȱtheȱtime,ȱitȱwasȱestimatedȱ
thatȱ overȱ 300ȱ energyȱ drinkȱ productȱ submissionsȱ wereȱ
beforeȱ Healthȱ Canadaȱ forȱ considerationȱ asȱ NHPs.ȱ Onlyȱ
twelveȱ wereȱ assessedȱ andȱ receivedȱ aȱ license.ȱ
Approximatelyȱ halfȱ ofȱ theȱ 300ȱ productȱ submissionsȱ wereȱ
eitherȱrefusedȱaȱlicenseȱorȱwithdrawnȱbyȱtheȱpetitioner.ȱAȱ
numberȱ ofȱ productsȱ wereȱ alsoȱ onȱ theȱ Canadianȱ marketȱ
underȱtheȱprovisionsȱofȱtheȱUnprocessedȱProductȱLicensingȱ
Regulations.ȱTableȱ1ȱbelowȱlistsȱtheȱmajorȱingredientsȱandȱ
levelsȱ typicallyȱ foundȱ inȱ theȱ productsȱ thatȱ wereȱ inȱ theȱ
queueȱ forȱ considerationȱ toȱ beȱ licensedȱ asȱ anȱ NHP.ȱ Forȱ
comparisonȱ purposes,ȱ Tableȱ 1ȱ alsoȱ providesȱ theȱ levelsȱ ofȱ
majorȱ ingredientsȱ thatȱ areȱ typicallyȱ foundȱ inȱ aȱ standardȱ
energyȱdrinkȱproduct.ȱȱȱ
standard,ȱ wereȱ rankedȱ nextȱ butȱ wellȱ belowȱ Redȱ Bullȱ
withȱ5.5%ȱandȱ4.4%ȱofȱtheȱmarketȱrespectively.ȱThisȱmarketȱ
shareȱ dataȱ hasȱ beenȱ evolvingȱ overȱ theȱ yearsȱ andȱ mayȱ notȱ
beȱreflectiveȱofȱtheȱcurrentȱmarketȱsituation.ȱȱ
ȱ
Aȱ wideȱ rangeȱ ofȱ energyȱ drinkȱ productsȱ isȱ availableȱ
internationallyȱ andȱ inȱ theȱ Canadianȱ marketplace.ȱ
Ingredientsȱ andȱ ingredientȱ levelsȱ varyȱ fromȱ productȱ toȱ
product.ȱ Whileȱ theȱ majorȱ ingredientsȱ typicallyȱ foundȱ inȱ
energyȱdrinksȱinȱCanadaȱareȱlistedȱinȱTableȱ1,ȱmanyȱotherȱ
ingredientsȱ canȱ beȱ foundȱ inȱ theseȱ products,ȱ particularlyȱ
variousȱherbsȱandȱextractsȱsuchȱasȱginseng,ȱginkgoȱbilobaȱ
andȱgreenȱteaȱextract.ȱȱ
2.3ȱȱEnergyȱdrinkȱmarketȱgrowthȱinȱCanadaȱ
Accordingȱ toȱ aȱ 2008ȱ reportȱ onȱ theȱ Canadianȱ energyȱ drinkȱ
marketȱ publishedȱ byȱ Agricultureȱ andȱ AgriȬFoodȱ Canadaȱ
(AAFC),ȱ inȱ 2006ȱ theȱ “functionalȱ beverageȱ category”ȱ wasȱ
valuedȱatȱ7%ȱofȱtheȱtotalȱsoftȱdrinkȱmarket.ȱEnergyȱdrinksȱ
areȱ includedȱ inȱ theȱ functionalȱ beverageȱ categoryȱ alongȱ
withȱ sportȱ andȱ nutraceuticalȱ drinks.ȱ Energyȱ drinksȱ
accountedȱ forȱ 65%ȱ ofȱ theȱ functionalȱ beverageȱ sectorȱ andȱ
4.6%ȱofȱtheȱvalueȱofȱtheȱsoftȱdrinkȱmarket.ȱInȱaȱmoreȱrecentȱ
reportȱ byȱ AAFCȱ (2011),ȱ itȱ wasȱ notedȱ thatȱ thereȱ areȱ moreȱ
thanȱ210ȱdifferentȱenergyȱdrinkȱ brandsȱinȱNorthȱAmerica.ȱ
Itȱ alsoȱ reportedȱ thatȱ theȱ energyȱ drinkȱ marketȱ grewȱ 60.2%ȱ
fromȱ2007ȱtoȱ2010ȱandȱ5.1%ȱfromȱ2009ȱtoȱ2010.ȱ
3.ȱHealthȱEffectsȱofȱEnergyȱDrinksȱ
ȱ
Substanceȱ
Formulationȱofȱproductsȱknownȱ
asȱEnergyȱDrinksȱthatȱhaveȱ
beenȱeitherȱlicensedȱorȱinȱqueueȱ
withȱHealthȱCanadaȱforȱ
considerationȱasȱNHPs.ȱ(mgȱperȱ
serving*)ȱ
Caffeineȱ
50ȱ–ȱ200ȱ
Taurineȱ
10ȱ–ȱ2000ȱ
Glucuronolactoneȱ
600ȱ–ȱ1200ȱ
niacinȱȱ
10ȱ–ȱ40ȱ
vitaminȱB6ȱȱ
5ȱ–ȱ10ȱ
vitaminȱB12ȱȱ
0.002ȱȬȱ0.2ȱȱȱȱ
pantothenicȱacidȱȱ
5ȱ–ȱ10ȱ
thiamineȱȱ
0.5ȱȬȱ5ȱȱȱȱȱ
riboflavinȱȱ
0.5ȱȬȱ5ȱȱȱ
Inositolȱ
50Ȭ200ȱ
Typicalȱ
energyȱ
drinkȱȱ
ȱ(mgȱperȱ
250ȱmlȱ
serving)ȱ
80ȱ
1000ȱ
600ȱ
18aȱ
2ȱ
0.001ȱ
6ȱ
2ȱ
1.65ȱ
50ȱ
*ȱServingȱsizeȱisȱtypicallyȱ250Ȭ473ȱml.ȱ
aȱEnergyȱ
drinksȱ containȱ niacinȱ eitherȱ inȱ theȱ formȱ ofȱ nicotinamideȱ orȱ
nicotinicȱacid.ȱRedȱBullȱcontainsȱnicotinamide.ȱ
Tableȱ1.ȱAmountsȱofȱmajorȱingredientsȱinȱenergyȱdrinksȱ
www.intechopen.com
ȱ
Energyȱ drinksȱ suchȱ asȱ Monsterȱ andȱ Rockstar,ȱ whichȱ
containȱ levelsȱ ofȱ ingredientsȱ higherȱ thanȱ theȱ standardȱ
and/orȱ containȱ otherȱ ingredientsȱ thatȱ areȱ notȱ partȱ ofȱ theȱ
3.1ȱȱIntroductionȱ
Theȱresearchȱonȱenergyȱdrinksȱisȱlargelyȱlimitedȱtoȱaȱsmallȱ
numberȱ ofȱ clinicalȱ studiesȱ (aboutȱ 12ȱ studies)ȱ andȱ caseȱ
reports.ȱTheȱclinicalȱstudiesȱtendedȱtoȱhaveȱsmallȱnumbersȱ
ofȱparticipantsȱ(lessȱthanȱ100ȱsubjects)ȱandȱfocusedȱonȱveryȱ
specificȱ healthȱ effects.ȱ Theseȱ studiesȱ mostlyȱ examinedȱ theȱ
effectȱ ofȱ energyȱ drinkȱ consumptionȱ onȱ behaviourȱ andȱ
physicalȱactivity;ȱtherefore,ȱtheȱparametersȱexaminedȱwereȱ
limitedȱ andȱ notȱ necessarilyȱ significantȱ fromȱ aȱ healthȱ andȱ
safetyȱperspective.ȱOtherȱstudiesȱassessedȱtheȱconsequencesȱ
ofȱconsumingȱtheȱcombinationȱofȱenergyȱdrinksȱandȱalcohol.ȱȱ
3.2ȱHealthȱeffectsȱofȱenergyȱdrinksȱȱ
3.2.1ȱAcuteȱphysiologicalȱeffectsȱofȱenergyȱdrinksȱ
Thereȱ isȱ concernȱ thatȱ someȱ individuals,ȱ whoȱ mayȱ haveȱ
increasedȱ sensitivityȱ toȱ theȱ ingredientsȱ inȱ energyȱ drinks,ȱ
Joel Rotstein, Jennifer Barber, Carl Strowbridge, Stephen Hayward, Rong Huang and Samuel Benrejeb Godefroy:
3
!
mayȱhaveȱanȱacuteȱphysiologicalȱresponse,ȱspecificallyȱanȱ
increaseȱ inȱ heartȱ rateȱ andȱ bloodȱ pressure.ȱ Tableȱ 2ȱ
summarizesȱtheȱlimitedȱstudiesȱonȱtheȱacuteȱphysiologicalȱ
effectsȱ ofȱ energyȱ drinks.ȱ Rashtiȱ etȱ al.ȱ (2009)ȱ investigatedȱ
10ȱhealthyȱwomen,ȱmeanȱageȱ20ȱyears.ȱSubjectsȱconsumedȱ
140ȱmlȱofȱanȱenergyȱdrinkȱorȱaȱplacebo.ȱAverageȱsystolicȱ
constitutesȱtheȱvastȱmajorityȱofȱcaloriesȱinȱtheseȱproducts.ȱ
Aȱ 250ȱ mlȱ servingȱ ofȱ aȱ typicalȱ energyȱ drinkȱ containsȱ 110ȱ
calories,ȱwhichȱisȱsimilarȱtoȱtheȱ86Ȭ130ȱcaloriesȱinȱtheȱsameȱ
volumeȱ ofȱ aȱ carbonatedȱ caffeinatedȱ softȱ drinkȱ (USDAȱ
NutrientȱDatabase,ȱ2011).ȱȱ
bloodȱ pressure1ȱ wasȱ significantlyȱ higherȱ forȱ theȱ energyȱ
drinkȱ groupȱ comparedȱ toȱ placebo.ȱ Noȱ differencesȱ wereȱ
seenȱ inȱ heartȱ rate,ȱ diastolicȱ bloodȱ pressureȱ orȱ profileȱ ofȱ
moodȱ statesȱ atȱ anyȱ timeȱ point.ȱ Americanȱ Heartȱ
Associationȱ (2007)ȱ reportedȱ onȱ aȱ studyȱ concerningȱ theȱ
cardiovascularȱeffectsȱofȱenergyȱdrinks.ȱFifteenȱsubjectsȱ(8ȱ
women,ȱ7ȱmen,ȱmeanȱageȱ26ȱyears)ȱwereȱgivenȱ500ȱmlȱofȱ
anȱ energyȱ drink.ȱ Theȱ researchersȱ notedȱ anȱ increaseȱ inȱ
systolicȱbloodȱpressureȱandȱanȱincreaseȱinȱheartȱrateȱinȱtheȱ
fourȱ hoursȱ afterȱ consumptionȱ ofȱ theȱ beverageȱ (citedȱ inȱ
BfRȱ 2008).ȱ However,ȱ acrossȱ theȱ variousȱ studies,ȱ changesȱ
inȱbloodȱpressureȱdidȱnotȱreachȱhypertensiveȱlevelsȱwithȱ
theȱ consumptionȱ ofȱ energyȱ drinks.ȱ Theseȱ effectsȱ wereȱ
similarȱ toȱ theȱ effectsȱ onȱ bloodȱ pressureȱ demonstratedȱ
withȱtheȱconsumptionȱofȱcoffeeȱ(Riksenȱetȱal.ȱ2009).ȱȱ
3.3ȱȱHealthȱEffectsȱofȱEnergyȱDrinksȱandȱSportsȱActivityȱ
3.2.2ȱBehavioural/performanceȱeffectsȱofȱenergyȱdrinksȱ
Aȱ limitedȱ numberȱ ofȱ studiesȱ haveȱ assessedȱ theȱ
behaviouralȱ effectsȱ followingȱ consumptionȱ ofȱ energyȱ
drinksȱ containingȱ bothȱ glucoseȱ andȱ caffeine,ȱ alongȱ withȱ
otherȱpotentiallyȱactiveȱagentsȱ(seeȱTableȱ3ȱbelow).ȱTheseȱ
studiesȱ haveȱ identifiedȱ improvementsȱ inȱ performanceȱ ofȱ
attentionȱ and/orȱ reactionȱ timeȱ tasksȱ andȱ variousȱ indicesȱ
ofȱ alertness.ȱ Forȱ example,ȱ Alfordȱ etȱ al.ȱ (2001)ȱ examinedȱ
theȱeffectsȱofȱRedȱBullȱEnergyȱDrinkȱoverȱ3ȱstudiesȱinȱ36ȱ
volunteers.ȱ Significantȱ improvementsȱ inȱ mentalȱ
performanceȱ
includingȱ
choiceȱ
reactionȱ
time,ȱ
concentrationȱandȱmemoryȱwereȱobservedȱwithȱRedȱBullȱ
comparedȱ toȱ theȱ controlȱ drinks.ȱ Scholeyȱ andȱ Kennedyȱ
(2004)ȱ investigatedȱ theȱ effectsȱ ofȱ glucoseȱ alone,ȱ caffeineȱ
alone,ȱginsengȱandȱginkgoȱatȱflavouringȱlevels,ȱanȱenergyȱ
drinkȱ orȱ aȱ placeboȱ inȱ 20ȱ students.ȱ Comparedȱ toȱ placebo,ȱ
theȱ energyȱ drinkȱ resultedȱ inȱ significantlyȱ improvedȱ
performanceȱ onȱ secondaryȱ memoryȱ andȱ speedȱ ofȱ
attentionȱ factors.ȱ Someȱ ofȱ theȱ studiesȱ attributedȱ theȱ
demonstratedȱeffectsȱtoȱtheȱcombinationȱofȱingredientsȱinȱ
energyȱ drinks,ȱ ratherȱ thanȱ caffeineȱ onlyȱ (Alfordȱ etȱ al.ȱ
2001;ȱReynerȱ&ȱHorneȱ2002;ȱScholeyȱ&ȱKennedyȱ2004).ȱ
3.2.3ȱCaloriesȱ
Exceptȱ forȱ sugarȬfreeȱ versions,ȱ energyȱ drinks,ȱ likeȱ manyȱ
beverages,ȱcontainȱsugarsȱinȱtheȱformȱofȱsucrose,ȱglucose,ȱ
and/orȱhighȬfructoseȱcornȱsyrup.ȱTheȱsugarȱcontentȱvariesȱ
amongȱenergyȱdrinksȱbutȱrangesȱfromȱ21ȱtoȱ34ȱgramsȱperȱ
237ȱ mlȱ canȱ (Clausonȱ etȱ al.ȱ 2008).ȱ Thisȱ sugarȱ contentȱ isȱ
similarȱ toȱ thatȱ ofȱ carbonatedȱ caffeinatedȱ softȱ drinks.ȱ Itȱ
Energyȱ drinksȱ areȱ frequentlyȱ marketedȱ toȱ individualsȱ
interestedȱ inȱ athleticsȱ andȱ anȱ activeȱ lifestyle.ȱ Theȱ mainȱ
ingredientsȱ inȱ energyȱ drinksȱ purportedȱ toȱ enhanceȱ sportȱ
performanceȱ areȱ caffeineȱ andȱ carbohydrates.ȱ Theȱ termȱ
“energyȱ drink”ȱ itselfȱ impliesȱ thatȱ itsȱ consumptionȱ mightȱ
enhanceȱphysicalȱactivity.ȱ
ȱ
Energyȱ drinksȱ shouldȱ notȱ beȱ confusedȱ withȱ ’sportsȱ
drinks’.ȱ Sportsȱ drinksȱ areȱ typicallyȱ aȱ mixtureȱ ofȱ
carbohydratesȱ andȱ electrolytesȱ formulatedȱ toȱ enhanceȱ
athleticȱ performanceȱ andȱ preventȱ dehydration,ȱ and,ȱ
unlikeȱ energyȱ drinks,ȱ theyȱ doȱ notȱ containȱ caffeine.ȱ
Conversely,ȱ energyȱ drinksȱ containȱ caffeineȱ whichȱ isȱ aȱ
stimulantȱ andȱ thereforeȱ theyȱ areȱ notȱ consideredȱ suitableȱ
toȱ reȬestablishȱ normalȱ bodyȱ functionȱ afterȱ exerciseȱ (e.g.ȱ
normalȱheartȱrate).ȱȱ
ȱ
Theȱ effectsȱ ofȱ energyȱ drinksȱ onȱ exerciseȱ haveȱ beenȱ
investigatedȱ inȱ aȱ smallȱ numberȱ ofȱ studiesȱ (seeȱ Tableȱ 4ȱ
below).ȱ Studiesȱ investigatingȱ theȱ useȱ ofȱ energyȱ drinksȱ
haveȱ generallyȱ foundȱ anȱ improvementȱ inȱ enduranceȱ
performance.ȱIvyȱetȱal.ȱ(2009)ȱfoundȱthatȱtheȱconsumptionȱ
ofȱ500ȱmlȱofȱaȱRedȱBullȱenergyȱdrinkȱ40ȱminutesȱbeforeȱaȱ
simulatedȱ cyclingȱ timeȱ trialȱ inȱ 12ȱ trainedȱ cyclistsȱ
significantlyȱimprovedȱenduranceȱperformanceȱcomparedȱ
toȱ aȱ nonȬcaffeinated,ȱ sugarȬfreeȱ placebo.ȱ Forbesȱ etȱ al.ȱ
(2007)ȱ foundȱ thatȱ theȱ consumptionȱ ofȱ Redȱ Bullȱ energyȱ
drinkȱ significantlyȱ increasedȱ upperȱ bodyȱ muscleȱ
enduranceȱbutȱhadȱnoȱeffectȱonȱanaerobicȱpeakȱorȱaverageȱ
powerȱ duringȱ repeatedȱ Wingateȱ cyclingȱ testsȱ inȱ healthyȱ
youngȱ adults,ȱ whenȱ comparedȱ toȱ aȱ nonȬcaffeinated,ȱ
isoenergetic,ȱ controlȱ beverage.ȱ Lockwoodȱ etȱ al.ȱ (2009)ȱ
foundȱ thatȱ preȬworkoutȱ energyȱ drinkȱ consumption,ȱ
significantlyȱimprovedȱsomeȱphysiologicalȱadaptationsȱtoȱ
combinedȱ aerobicȱ andȱ resistanceȱ training,ȱ whenȱ
comparedȱ toȱ aȱ nonȬcaffeinated,ȱ sugarȬfreeȱ controlȱ drink.ȱ
Lastly,ȱ aȱ studyȱ investigatingȱ theȱ useȱ ofȱ aȱ sugarȬfreeȱ
energyȱ drinkȱ ȱ (Redȱ Bull)ȱ foundȱ thatȱ thereȱ wasȱ noȱ
differenceȱinȱrunȱtimeȬtoȬexhaustionȱorȱperceivedȱexertionȱ
inȱ youngȱ adultsȱ whenȱ comparedȱ toȱ nonȬcaffeinatedȱ
sugarȬfreeȱ placeboȱ (Candowȱ etȱ al.ȱ 2009).ȱ Overallȱ theȱ
studies’ȱ resultȱ suggestȱ thatȱ theȱ consumptionȱ ofȱ energyȱ
drinksȱ mayȱ enhanceȱ physicalȱ abilityȱ butȱ itȱ isȱ notȱ clearȱ
whetherȱtheȱeffectȱisȱdueȱtoȱtheȱpresenceȱofȱcaffeine,ȱsugarȱ
orȱbothȱofȱtheseȱconstituentsȱinȱtheȱenergyȱdrink.ȱ
ȱ
ȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱ ȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱ
ȱDuringȱ eachȱ heartbeat,ȱ bloodȱ pressureȱ variesȱ betweenȱ aȱ
maximumȱ(systolic)ȱandȱaȱminimumȱ(diastolic)ȱpressure.ȱ
1
4
ȱ
www.intechopen.com
!
StudyȱDesignȱ
DoubleȬblind,ȱ
placeboȬcontrolledȱ
Randomized,ȱ
doubleȬblind,ȱ crossȬ
over,ȱ
placeboȬ
controlledȱ
Prospectiveȱ
PlaceboȬcontrolledȱ
TestȱMaterialȱ
(composition)ȱ1ȱ
RedȱBullȱEnergyȱDrinkȱ
(32ȱmg/dLȱcaffeine,ȱ240ȱmg/dLȱ
glucuronolactone,ȱ400ȱmg/dlȱtaurine)ȱ
ȱ
LowȱCalorieȱRedȱBullȱEnergyȱDrinkȱ(32ȱ
mg/dLȱcaffeine,ȱ240ȱmg/dLȱ
glucuronolactone,ȱ400ȱmg/dlȱtaurine)ȱ
ȱ
HighȱCalorieȱPlaceboȱ(carbonatedȱwater,ȱ
tapȱwaterȱandȱdextrose)ȱ
ȱ
LowȱCalorieȱPlaceboȱ(carbonatedȱwater,ȱ
tapȱwaterȱandȱdextrose)ȱ
ȱ
NB:ȱBananaȱflavouringȱandȱgreenȱ
colouringȱwereȱaddedȱtoȱallȱtestȱmaterials
Methods
Studyȱ1:ȱ69ȱsubjectsȱ
(48ȱwomen,ȱ21ȱ
males,ȱmeanȱageȱ20ȱ
y)ȱconsumedȱ250ȱmlȱ
RedȱBull,ȱLowȱ
CalorieȱRedȱBull,ȱ
highȱcalorieȱ
placebo,ȱorȱlowȱ
calorieȱplacebo.ȱ
ȱ
Studyȱ2:ȱ21ȱsubjectsȱ
(9ȱmen,ȱ12ȱwomen,ȱ
meanȱageȱ22ȱy)ȱ
consumedȱ250ȱmlȱ
RedȱBullȱpreȱandȱ
postȱcoldȱpressorȱ
test.ȱ
10ȱsubjectsȱ(meanȱ
ageȱ20ȱy)ȱreceivedȱ
140ȱmlȱofȱMeltdownȱ
RTDȱenergyȱdrinkȱ
orȱaȱplacebo.ȱ
MeltdownȱRTDȱ
(140ȱmlȱcontains:ȱ230ȱmgȱcaffeine,ȱ
unknownȱamountsȱofȱ
methyltetradecythioaceticȱacid,ȱyerbaȱ
mateȱextract,ȱmethylȬsynephrine,ȱ
methylphenylethylene,ȱ11Ȭhydroxyȱȱ
yohimbine,ȱyohimbineȱHCL,ȱalphaȬ
yohimbiine,ȱandȱmethylȬhordenineȱHCl)ȱ
ȱ
Placeboȱ
(describedȱasȱsimilarȱinȱappearanceȱandȱ
tasteȱtoȱMeltdownȱRTDȱbutȱonlyȱ
containedȱinertȱsubstances)ȱ
15ȱsubjectsȱ(7ȱmen,ȱ8ȱ
Unspecifiedȱenergyȱdrinkȱ(250ȱmlȱ
women,ȱmeanȱageȱ
contains:ȱ1000ȱmgȱtaurine,ȱ100ȱmgȱ
caffeine,ȱunstatedȱamountsȱofȱvitaminsȱB5,ȱ 26ȱy)ȱconsumedȱ500ȱ
mlȱ(2ȱcans)ȱofȱanȱ
B6,ȱB12,ȱglucuronolactone,ȱniacinamide)ȱ
energyȱdrinkȱdailyȱ
forȱtheȱnextȱ5ȱdays.
Resultsȱ
Reference
Studyȱ1:ȱNoȱchangesȱnotedȱ Ragsdaleȱetȱal.ȱ
2010ȱ
inȱoverallȱcardiovascularȱ
functionȱorȱbloodȱglucoseȱ
overȱ2ȱhourȱtestȱperiod.ȱ
ȱ
ȱ
Studyȱ2:ȱAȱsignificantȱ
increaseȱinȱdiastolicȱbloodȱ
pressureȱinȱtheȱmalesȱ
immediatelyȱafterȱ
submersionȱofȱtheȱhandȱinȱ
coldȱwaterȱ(5oȱC).ȱNoȱ
significantȱchangeȱnotedȱinȱ
females.ȱ
Meanȱsystolicȱbloodȱ
pressureȱwasȱsignificantlyȱ
higherȱforȱtheȱenergyȱdrinkȱ
groupȱcomparedȱtoȱplacebo.ȱ
Noȱdifferencesȱnotedȱinȱ
heartȱrate,ȱdiastolicȱbloodȱ
pressure,ȱorȱprofileȱofȱmoodȱ
states.ȱ
Rashtiȱetȱal.ȱ
2009ȱ
Steinkeȱetȱal.ȱ
Withinȱ4ȱhoursȱofȱenergyȱ
2009ȱ
drinkȱconsumption,ȱmeanȱ
systolicȱbloodȱpressureȱandȱ
heartȱrateȱsignificantlyȱ
increased.ȱDiastolicȱbloodȱ
pressureȱsignificantlyȱ
increasedȱwithinȱ2ȱhoursȱofȱ
energyȱdrinkȱconsumption.ȱ
50ȱsubjectsȱ(34ȱmen,ȱ Comparedȱwithȱbaselineȱ Worthleyȱetȱal.ȱ
2010ȱ
values,ȱthereȱwasȱaȱ
contains:ȱ1000ȱmgȱtaurine,ȱ80ȱmgȱcaffeine,ȱ 16ȱwomen,ȱmeanȱ
significantȱincreaseȱinȱ
ageȱ22ȱy)ȱconsumedȱ
600ȱmgȱglucuronolcatone,ȱunstatedȱ
plateletȱaggregationȱ
eitherȱaȱ250ȱmlȱ
amountsȱofȱȱvitaminsȱB2,ȱB5,ȱB6ȱandȱB12,ȱ
followingȱenergyȱdrinkȱ
sugarȬfreeȱenergyȱ
sugarȬfreeȱsweetenerȱandȱthickeners)ȱ
consumption;ȱnoȱchangeȱ
drinkȱorȱ250ȱmlȱofȱ
ȱ
carbonatedȱwaterȱ wasȱobservedȱwithȱcontrol.ȱ
Placeboȱ
Meanȱarterialȱpressureȱ
(control)ȱ
(carbonatedȱwater)ȱ
significantlyȱincreasedȱ
followingȱenergyȱdrinkȱ
consumptionȱcomparedȱwithȱ
control.ȱ
1ȱTestȱmaterialȱcompositionȱisȱlistedȱasȱitȱisȱdescribedȱinȱtheȱpublishedȱreport.ȱ
Tableȱ2.ȱStudiesȱinvestigatingȱtheȱacuteȱphysiologicalȱeffectsȱofȱenergyȱdrinksȱ
ȱ
ȱ
ȱ
ȱ
ȱ
ȱ
ȱ
www.intechopen.com
ȱ
5
!
StudyȱDesignȱ
Testȱmaterialȱ
Methods
Resultsȱ
Reference
(composition)1ȱ
Alfordȱetȱal.ȱ
Comparedȱtoȱcontrolȱ
Randomized,ȱ
RedȱBullȱEnergyȱDrinkȱ
2001ȱ
doubleȬblindȱ
(250ȱmlȱcontains:ȱcarbonatedȱwater,ȱ women,ȱageȱrangeȱ18Ȭ30ȱy)ȱinȱ drinks,ȱRedȱBullȱenergyȱ
drinkȱsignificantlyȱ
21.5ȱgȱsucrose,ȱ5.25ȱglucose,ȱ1000ȱmgȱ 3ȱstudiesȱreceivedȱnoȱdrink,ȱ
improvedȱmentalȱ
taurine,ȱ600ȱmgȱglucuronolactone,ȱ 250ȱmlȱofȱcarbonatedȱwater,ȱ
RedȱBullȱenergyȱȱdrink,ȱorȱ performanceȱincludingȱ
80ȱmgȱcaffeine,ȱ50ȱmgȱinositiol,ȱ
choiceȱreactionȱtime,ȱ
“dummy”ȱenergyȱdrink.ȱ
unstatedȱamountsȱofȱniacin,ȱ
concentration,ȱandȱ
panthenol,ȱB6,ȱB12,ȱȱriboflavin,ȱ
memory.ȱ
flavours,ȱcolours)ȱ
ȱ
Carbonatedȱwaterȱ
ȱ
Dummyȱenergyȱdrinkȱ(unstatedȱ
amountsȱofȱlowȱcalorieȱquinineȱ
flavouredȱcarbonatedȱwaterȱwithȱ
lime,ȱappleȱandȱblackcurrantȱ
concentrates)ȱ
ȱ
Theȱenergyȱdrinkȱ
11ȱsubjectsȱ(maleȱandȱfemale,ȱ
DoubleȬblindȱ
RedȱBullȱEnergyȱDrinkȱ(250ȱmlȱ
Horneȱ&ȱ
contains:ȱ21ȱgȱsucrose,ȱ5ȱgȱglucose,ȱ1ȱ meanȱageȱ24ȱy)ȱreceivedȱ500ȱ significantlyȱreducedȱ
Reynerȱ2001ȱ
laneȱdriftingȱandȱ
gȱtaurine,ȱ600ȱmgȱglucuronolactone,ȱ mlȱofȱanȱenergyȱdrinkȱorȱaȱ
improvedȱreactionȱ
controlȱdrinkȱpriorȱtoȱcarȱ
80ȱmgȱcaffeine,ȱ50ȱmgȱinositol,ȱandȱ
simulatorȱtest.ȱControlȱdrinkȱ time,ȱparticularlyȱforȱ
unstatedȱamountsȱofȱvitaminȱBȱ
theȱ1stȱhour.ȱ
consistedȱofȱidenticalȱdrinkȱ
complex)ȱ
minusȱtheȱcaffeine,ȱtaurineȱ
ȱ
andȱglucuronolactone.ȱ
Placeboȱ(250ȱmlȱcontains:ȱ21ȱgȱ
sucrose,ȱ5ȱgȱglucose,ȱȱ50ȱmgȱinositol,ȱ
andȱunstatedȱamountsȱofȱvitaminȱBȱ
complex)ȱ
12ȱsubjectsȱ(7ȱmen,ȱ5ȱwomen,ȱ Comparedȱwithȱtheȱ
Reynerȱ&ȱ
DoubleȬblindȱ
control,ȱtheȱenergyȱ
Horneȱ2002ȱ
contains:ȱ21ȱgȱsucrose,ȱ5ȱgȱglucose,ȱ1ȱ meanȱageȱ24ȱy)ȱreceivedȱ250ȱ
gȱtaurine,ȱ600ȱmgȱglucuronolactone,ȱ mlȱofȱRedȱBullȱenergyȱdrinkȱorȱ drinkȱsignificantlyȱ
aȱcontrolȱdrinkȱpriorȱtoȱcarȱ reducedȱsleepȬrelatedȱ
80ȱmgȱcaffeine,ȱ50ȱmgȱinositol,ȱandȱ
simulatorȱtest.ȱControlȱdrinkȱ drivingȱincidentsȱandȱ
unstatedȱamountsȱofȱvitaminȱBȱ
subjectiveȱsleepinessȱ
consistedȱofȱidenticalȱdrinkȱ
complex)ȱ
minusȱtheȱcaffeine,ȱtaurineȱ forȱtheȱ1stȱ90ȱminutesȱofȱ
ȱ
andȱglucuronolactone.ȱ
Placeboȱ(250ȱmlȱcontains:ȱ21ȱgȱ
theȱdrive.ȱ
sucrose,ȱ5ȱgȱglucose,ȱȱ50ȱmgȱinositol,ȱ
andȱunstatedȱamountsȱofȱvitaminȱBȱ
complex)ȱ
ȱ
Scholeyȱ&ȱ
Comparedȱwithȱ
Placeboȱ(250ȱmlȱcontains:ȱwaterȱwithȱ 20ȱsubjectsȱ(14ȱwomen,ȱ6ȱmen,ȱ
Randomized,ȱ
Kennedyȱ2004ȱ
meanȱageȱ21ȱy)ȱreceivedȱ1ȱofȱ5ȱ placebo,ȱtheȱenergyȱ
flavourȱandȱsweeteners),ȱ
doubleȬblind,ȱ5ȱ wayȱ
drinkȱresultedȱinȱ
drinks:ȱplacebo,ȱdrinkȱ1,ȱdrinkȱ
ȱ
crossȬover,ȱ placeboȬ
Drinkȱ1ȱ(250ȱmlȱcontains:ȱwaterȱwithȱ 2,ȱdrinkȱ3ȱorȱanȱenergyȱdrinkȱ significantlyȱimprovedȱ
controlledȱ
performanceȱonȱ
forȱaȱtotalȱofȱ6ȱstudyȱdays,ȱ
flavour,ȱsweetenersȱandȱ75ȱmgȱ
secondaryȱmemoryȱandȱ
eachȱ7ȱdaysȱapart.ȱ
caffeineȱ
speedȱofȱattentionȱ
ȱ
factors.ȱ
Drinkȱ2ȱ(250ȱml:ȱwaterȱwithȱflavour,ȱ
sweetenersȱandȱ37.5ȱgȱglucose)ȱ
ȱ
Drinkȱ3ȱ(250ȱmlȱcontains:ȱwaterȱwithȱ
flavour,ȱsweeteners,ȱ12.5ȱmgȱginsengȱ
extract,ȱ2ȱmgȱginkgoȱbilobaȱextract)
ȱ
contains:ȱwaterȱwithȱflavour,ȱ
sweeteners,ȱ75ȱmgȱcaffeine,ȱ37.5ȱgȱ
glucose,ȱ12.5ȱmgȱginsengȱextract,ȱ2ȱ
mgȱginkgoȱbilobaȱextract)ȱ
ȱ
ȱ
6
ȱ
www.intechopen.com
!
Warburtonȱetȱ
Theȱenergyȱdrinkȱ
42ȱsubjectsȱ(ageȱ18Ȭ24ȱy)ȱ
DoubleȬblind,ȱ2ȱwayȱ Unspecifiedȱenergyȱdrinkȱ(250ȱmlȱ
al.ȱ2001ȱ
crossȬover,ȱ placeboȬ contains:ȱ80ȱmgȱcaffeine,ȱ1000ȱmgȱ consumedȱ250ȱmlȱofȱanȱenergyȱ improvedȱattentionȱandȱ
verbalȱreasoningȱ
drinkȱorȱaȱplacebo,ȱeitherȱaȱ
taurine,ȱ600ȱmgȱglucuronolactone,ȱ
controlledȱ
comparedȱtoȱtheȱsugarȬ
sugarȬfreeȱplaceboȱorȱaȱ
5.25ȱmgȱglucose,ȱ21.ȱ5ȱmgȱsucrose,ȱ50ȱ
freeȱandȱtheȱadditionalȱ
placeboȱwithȱadditionalȱ
mgȱinositol,ȱ20ȱmgȱniacin,ȱ5ȱmgȱ
sugarȬcontainingȱ
glucose.ȱ
vitaminȱB6,ȱ5ȱmgȱvitaminȱB5,ȱ1.5ȱmgȱ
placebos.ȱ
vitaminȱB2,ȱandȱ0.005ȱmgȱvitaminȱ
B12)ȱ
ȱ
SugarȬfreeȱplaceboȱ(250ȱmlȱcontains:ȱ
1000ȱmgȱtaurine,ȱ600ȱmgȱ
glucuronolactone,5.25ȱmgȱglucoseȱ50ȱ
mgȱinositol,ȱ20ȱmgȱniacin,ȱ5ȱmgȱ
vitaminȱB6,ȱ5ȱmgȱvitaminȱB5,ȱ1.5ȱmgȱ
vitaminȱB2,ȱandȱ0.005ȱmgȱvitaminȱ
B12)ȱ
ȱ
Placeboȱ(250ȱmlȱcontains:ȱ22.5ȱmgȱ
caffeine,ȱ1000ȱmgȱtaurine,ȱ600ȱmgȱ
glucuronolactone,ȱ6.5ȱmgȱglucose,ȱ
21.5ȱmgȱsucrose,ȱ50ȱmgȱinositol,ȱ20ȱ
mgȱniacin,ȱ5ȱmgȱvitaminȱB6,ȱ5ȱmgȱ
vitaminȱB5,ȱ1.5ȱmgȱvitaminȱB2,ȱandȱ
0.005ȱmgȱvitaminȱB12)ȱ
Tableȱ3.ȱStudiesȱinvestigatingȱbehavioural/performanceȱeffectsȱofȱenergyȱdrinksȱȱ
StudyȱDesignȱ
Testȱmaterialȱ
Methods
Resultsȱ
Reference
(composition)ȱ1ȱ
Candowȱetȱal.ȱ
Noȱsignificantȱ
17ȱsubjectsȱȱ(9ȱmen,ȱ8ȱwomen,ȱ
SugarȬfreeȱRedȱ
Randomized,ȱ
2009ȱ
differenceȱinȱrunȱtimeȬ
meanȱageȱ21ȱy)ȱreceivedȱ
doubleȬblind,ȱ crossȬ BullȱEnergyȱDrinkȱ(2ȱmg/kgȱcaffeine;ȱ
toȬexhaustion,ȱ
eitherȱsugarȬfreeȱRedȱBullȱ
~146ȱmgȱcaffeine)ȱ
over,ȱ
placeboȬ
perceivedȱexertionȱorȱ
energyȱdrinkȱorȱ
ȱ
controlledȱ
decaffeinated,ȱsugarȬfreeȱ calciumȱlactateȱbetweenȱ
Placeboȱ(decaffeinated,ȱsugarȬfree,ȱ
groups.ȱ
placeboȱonȱ3ȱdaysȱofȱtesting,ȱ
lemonȬlimeȱflavouredȱsoftȱdrink,ȱ
duringȱwhichȱtheyȱ
tonicȱwater,ȱandȱlimeȱjuice)ȱ
performedȱphysicalȱactivity.
ȱ
RedȱBullȱEnergyȱDrinkȱ Forbesȱetȱal.ȱ
RedȱBullȱEnergyȱDrinkȱ(2ȱmg/kgȱ
Randomized,ȱ
2007ȱ
significantlyȱincreasedȱ
women,ȱmeanȱageȱ21ȱy)ȱ
caffeine;ȱ~146ȱmgȱcaffeine)ȱ
upperȱbodyȱmuscleȱ
receivedȱRedȱBullȱEnergyȱ
ȱ
over,ȱ
placeboȬ
Placeboȱ(isoenergetic,ȱisovolumetric,ȱ Drinkȱ(2ȱmg/kgȱcaffeine)ȱorȱ enduranceȱbutȱhadȱnoȱ
controlledȱ
effectȱonȱanaerobicȱ
nonȬcaffeinatedȱplaceboȱ
nonȱcaffeinated)ȱ
separatedȱbyȱ7ȱdays.ȱMuscleȱ peakȱorȱaverageȱpower.ȱ
ȱ
enduranceȱassessedȱbyȱtheȱ
ȱ
maximumȱnumberȱofȱ
ȱ
repetitionsȱoverȱ3ȱsets.ȱThreeȱ
Wingateȱcyclingȱtestsȱwereȱ
usedȱtoȱassessȱpeakȱandȱ
averageȱpowerȱoutput.ȱ
ȱ
ȱ
Ivyȱetȱal.ȱ2009ȱ
Performanceȱ
12ȱtrainedȱcyclistsȱ(6ȱmen,ȱ6ȱ
Randomized,ȱ
significantlyȱimprovedȱ
women,ȱmeanȱageȱ27ȱy)ȱ
contains:ȱ2ȱgȱtaurine,ȱ1.2ȱgȱ
withȱenergyȱdrinkȱ
over,ȱ
placeboȱ glucuronolactone,ȱ160ȱmgȱcaffeine,ȱ54ȱ consumedȱ500ȱmlȱofȱeitherȱ
gȱcarbohydrate,ȱ40ȱmgȱniacin,ȱ10ȱmgȱ placeboȱorȱRedȱBullȱEnergyȱ comparedȱtoȱplaceboȱ
controlledȱ
butȱnoȱdifferenceȱinȱ
Drinkȱ40ȱminutesȱbeforeȱ
pantothenicȱacid,ȱ10ȱmgȱvitaminȱB6,ȱ
ratingȱofȱperceivedȱ
simulatedȱcyclingȱtimeȱtrial.ȱ
10ȱugȱvitaminȱB12)ȱ
exertionȱbetweenȱ
ȱ
treatments.ȱ
Placeboȱ(500ȱmlȱcontains:ȱflavouredȱ
drink)ȱ
ȱ
ȱ
www.intechopen.com
ȱ
7
!
Randomized,ȱ
doubleȬblind,ȱ
placeboȬcontrolledȱ
38ȱsubjectsȱ(male,ȱ18ȱȬ45ȱ
Unspecifiedȱenergyȱdrinkȱ(12ȱflȱozȱ
yearsȱold)ȱreceivedȱenergyȱ
contains:ȱcarbonatedȱwater,ȱcitricȱ
drinkȱ+ȱexercise,ȱenergyȱ
acid,ȱnaturalȱflavoursȱ(lemonȬlime),ȱ
sucralose,ȱ60ȱmgȱvitaminȱC,ȱ1.7ȱmgȱ drink,ȱplacebo,ȱorȱplaceboȱ+ȱ
exerciseȱforȱ10ȱweeksȱ(1ȱdrinkȱ
riboflavin,ȱ20ȱmgȱniacin,ȱ2ȱmgȱ
perȱday)ȱ
vitaminȱB6,ȱ6ȱugȱvitaminȱB12,ȱ300ȱugȱ
biotin,ȱ10ȱmgȱpantothenicȱacid,ȱ50ȱmgȱ
calcium,ȱ50ȱugȱchromium,ȱ6ȱmgȱ
sodium,ȱ1810ȱmgȱcombinedȱtaurine,ȱ
guaranaȱextract,ȱgreenteaȱleafȱextract,ȱ
caffeine,ȱglucuronolactone,ȱgingerȱ
extractȱ(totalȱ200ȱmgȱcaffeine))ȱ
ȱ
Placeboȱ(12ȱflȱozȱcontains:ȱcarbonatedȱ
water,ȱcitricȱacid,ȱnaturalȱflavoursȱ
(lemonȬlime),ȱsucralose)ȱ
Significantlyȱgreaterȱ
decreasesȱinȱfatȱmassȱ
andȱpercentageȱbodyȱ
fat,ȱasȱwellȱasȱpowerȱ
outputȱatȱventilatoryȱ
thresholdȱwereȱ
observedȱinȱenergyȱ
drinkȱ+ȱexerciseȱ
comparedȱtoȱplaceboȱ+ȱ
exercise.ȱ
Lockwoodȱetȱ
al.ȱ2009ȱ
Tableȱ4.ȱStudiesȱinvestigatingȱenergyȱdrinksȱandȱexercise
3.4ȱȱEnergyȱDrinksȱandȱAlcoholȱȱ
Dueȱ toȱ concernsȱ withȱ theȱ combinationȱ ofȱ energyȱ drinksȱ
andȱalcohol,ȱHealthȱCanadaȱhasȱrecommendedȱsinceȱ2004,ȱ
theȱdateȱofȱlicensingȱofȱtheȱfirstȱEnergyȱDrinksȱasȱNHPs,ȱ
thatȱ theseȱ productsȱ ȱ shouldȱ notȱ beȱ mixedȱ withȱ alcoholȱ
(HealthȱCanada:ȱIt’sȱYourȱHealthȱ2010).ȱNonetheless,ȱtheȱ
ingestionȱ ofȱ energyȱ drinksȱ withȱ alcoholȱ hasȱ becomeȱ
increasinglyȱ popularȱ (Healthȱ Canada:ȱ It’sȱ Yourȱ Healthȱ
2010).ȱ Aȱ surveyȱ ofȱ 496ȱ Americanȱ collegeȱ studentsȱ
examinedȱ theȱ frequencyȱ ofȱ energyȱ drinkȱ useȱ forȱ sixȱ
“situations”:ȱ insufficientȱ sleep,ȱ toȱ increaseȱ energy,ȱ whileȱ
studying,ȱ drivingȱ longȱ periodsȱ ofȱ time,ȱ drinkingȱ withȱ
alcohol,ȱ andȱ toȱ treatȱ hangover.ȱ Theȱ majorityȱ ofȱ usersȱ
consumedȱ oneȱ energyȱ drinkȱ toȱ treatȱ mostȱ situationsȱ
(namely,ȱ hangover,ȱ insufficientȱ sleep,ȱ increaseȱ energy,ȱ
drivingȱ aȱ carȱ forȱ aȱ longȱ periodȱ ofȱ time)ȱ althoughȱ usingȱ
threeȱorȱmoreȱtoȱdrinkȱwithȱalcoholȱwhileȱpartyingȱwasȱaȱ
commonȱ practiceȱ (49%)ȱ (Malinauskasȱ etȱ al.ȱ 2007).ȱ Inȱ aȱ
surveyȱ ofȱ 4271ȱ Americanȱ collegeȱ students,ȱ 24%ȱ (697ȱ
students)ȱreportedȱconsumingȱenergyȱdrinksȱwithȱalcoholȱ
(O’Brienȱetȱal.ȱ2008).ȱAȱsummaryȱofȱstudiesȱinvestigatingȱ
theȱ useȱ ofȱ energyȱ drinksȱ andȱ alcoholȱ canȱ beȱ foundȱ inȱ
Tableȱ5ȱbelow.ȱ
ȱ
Basicȱ physiologyȱ suggestsȱ thatȱ theȱ carbonationȱ ofȱ theȱ
energyȱ drinkȱ permitsȱ aȱ quickerȱ absorptionȱ ofȱ alcoholȱ
whileȱ theȱ caffeineȱ ofȱ theȱ energyȱ drinkȱ mayȱ maskȱ theȱ
drowsinessȱ associatedȱ withȱ alcoholȱ intake.ȱ Althoughȱ
thereȱ isȱ littleȱ directȱ evidenceȱ forȱ it,ȱ whenȱ mixingȱ energyȱ
drinksȱ andȱ alcohol,ȱ usersȱ mayȱ notȱ feelȱ theȱ symptomsȱ ofȱ
alcoholȱintoxicationȱwhichȱmayȱincreaseȱtheȱpotentialȱforȱ
alcoholȬrelatedȱ injuryȱ (Ferreiraȱ etȱ al.,ȱ 2006)2.ȱ Aȱ surveyȱ ofȱ
AmericanȱcollegeȱstudentsȱbyȱO’Brienȱetȱal.ȱ(2008)ȱfoundȱ
thatȱinȱcomparisonȱtoȱthoseȱwhoȱconsumedȱalcoholȱalone,ȱ
studentsȱ whoȱ consumedȱ alcoholȱ mixedȱ withȱ energyȱ
drinksȱ hadȱ aȱ significantlyȱ higherȱ prevalenceȱ ofȱ alcoholȬ
relatedȱ consequences,ȱ including:ȱ beingȱ takenȱ advantageȱ
of,ȱ orȱ takingȱ advantageȱ ofȱ anotherȱ studentȱ sexually,ȱ
ridingȱinȱanȱautomobileȱwithȱaȱdriverȱunderȱtheȱinfluenceȱ
ofȱ alcohol,ȱ orȱ beingȱ hurtȱ orȱ injured.ȱ Inȱ addition,ȱ mixingȱ
energyȱdrinksȱwithȱalcoholȱwasȱassociatedȱwithȱincreasedȱ
heavyȱ episodicȱ drinkingȱ andȱ episodesȱ ofȱ weeklyȱ
drunkenness.ȱ Thombsȱ etȱ al.ȱ (2010)ȱ conductedȱ aȱ fieldȱ
studyȱ inȱ aȱ U.S.ȱ barȱ district,ȱ interviewingȱ 802ȱ randomlyȱ
selectedȱ andȱ selfȬselectedȱ participantsȱ andȱ foundȱ thatȱ
thoseȱwhoȱmixedȱenergyȱdrinksȱwithȱalcoholȱwereȱatȱaȱ3Ȭ
foldȱriskȱofȱleavingȱaȱbarȱhighlyȱintoxicatedȱasȱwellȱasȱaȱ4Ȭ
foldȱ riskȱ ofȱ intendingȱ toȱ driveȱ uponȱ leavingȱ theȱ barȱ
district,ȱcomparedȱtoȱthoseȱwhoȱdidȱnotȱmixȱalcoholȱwithȱ
energyȱdrinks.ȱ
ȱ
Inȱ Germany,ȱ theȱ Federalȱ Instituteȱ forȱ Riskȱ Assessmentȱ
(BfR)ȱ recommendsȱ thatȱ “adverseȱ effectsȱ cannotȱ beȱ ruledȱ
outȱ whenȱ largerȱ amountsȱ ofȱ theseȱ beveragesȱ areȱ
consumedȱinȱconjunctionȱwithȱintensiveȱphysicalȱactivityȱ
orȱwithȱtheȱintakeȱofȱalcoholȱbeverages”.ȱTheȱBfRȱlookedȱ
atȱcaseȱreportsȱfromȱtheirȱSwedishȱregulatoryȱcounterpart,ȱ
theȱ Nationalȱ Foodȱ Administration.ȱ Informationȱ obtainedȱ
fromȱtheȱSwedishȱliteratureȱsearchȱisȱdescribedȱinȱTableȱ6ȱ
below.ȱ
ȱ
ȱWhileȱ thisȱ manuscriptȱ wasȱ inȱ preparation,ȱ aȱ studyȱ wasȱ
publishedȱwhichȱsuggestedȱthatȱenergyȱdrinksȱdidȱnotȱmaskȱtheȱ
effectȱ ofȱ alcoholȱ (Alfordȱ etȱ al.,ȱ 2012).ȱ Asȱ well,ȱ aȱ statementȱ wasȱ
publishedȱ byȱ theȱ Committeeȱ onȱ Toxicity(UK)ȱ 2012,ȱ whichȱ
concludedȱthatȱtheȱlimitedȱcurrentlyȱavailableȱevidenceȱdoesȱnotȱ
supportȱaȱharmfulȱinteractionȱbetweenȱalcoholȱandȱcaffeine.ȱ
2
8
ȱ
www.intechopen.com
!
ȱ
StudyȱDesignȱ
Testȱmaterialȱ1ȱ
Methods
Results
Study
Curryȱandȱ
GreenȱMonsterȱ(16ȱozȱ 27ȱfemaleȱsubjectsȱ(meanȱ Energyȱdrinkȱplusȱalcoholȱsignificantlyȱloweredȱ
Randomized,ȱ
Stasioȱ2009ȱ
postȬtestȱperformanceȱonȱaȱglobalȱscoreȱofȱ
contains:ȱ160ȱmgȱ age:ȱ22ȱyears)ȱconsumedȱ
doubleȬblind,ȱ
neuropsychologicalȱstatus,ȱspecificallyȱ
caffeine,ȱ52ȱgȱsugar,ȱ GreenȱMonsterȱenergyȱ
placeboȬ
visuospatial/ȱconstructionalȱandȱlanguageȱ
2000ȱmgȱtaurine,ȱ400ȱ drinkȱaloneȱ(1ȱdrink),ȱ
controlledȱ
performanceȱscores.ȱ
mgȱginseng,ȱunstatedȱ withȱalcohol,ȱorȱaȱnonȬ
alcoholic,ȱnonȬ
amountȱofȱguarana)ȱ
caffeinatedȱcontrol.ȱ
ȱ
Alcoholic,ȱ
caffeinatedȱȱbeverageȱ
(16ȱozȱcontainsȱ:ȱȱ6%ȱ
alcohol,ȱ87ȱmgȱ
caffeine,ȱ47ȱgȱsugar,ȱ
unstatedȱamountsȱofȱ
taurine,ȱginsengȱandȱ
guarana)ȱ
ȱ
NonȬalcoholic,ȱnonȬ
caffeinatedȱbeverageȱ
(16ȱozȱcontainsȱ:ȱ5ȱgȱ
sugar)ȱ
Randomized,ȱ
RedȱBullȱenergyȱ
26ȱmaleȱsubjectsȱ(meanȱ Whenȱcomparedȱtoȱtheȱingestionȱofȱalcoholȱalone,ȱ Ferreiraȱetȱal.ȱ
2006ȱ
doubleȬblindȱ
drinkȱ(250ȱml/can)ȱ ageȱ23ȱy)ȱreceivedȱRedȱ theȱingestionȱofȱalcoholȱ+ȱenergyȱdrinkȱsignificantlyȱ
reducedȱsubjects’ȱperceptionȱofȱheadache,ȱ
BullȱEnergyȱdrinkȱ(1ȱ
weakness,ȱdryȱmouth,ȱandȱimpairmentȱofȱmotorȱ
can)ȱplusȱvodkaȱ(0.6ȱ
coordination.ȱ
g/kgȱorȱ1ȱg/kg)ȱorȱ
alcoholȱaloneȱorȱenergyȱ
drinkȱalone.ȱ
Unspecifiedȱenergyȱ 496ȱUSȱcollegeȱstudentsȱ Theȱmajorityȱofȱusersȱreportedȱconsumingȱenergyȱ Malinauskasȱ
etȱal.ȱ2007ȱ
drinksȱ
completedȱ
drinksȱforȱinsufficientȱsleepȱ(67%),ȱtoȱincreaseȱ
Surveyȱandȱfieldȱ
questionnairesȱ
energyȱ(65%),ȱandȱtoȱdrinkȱalcoholȱwhileȱpartyingȱ
testȱ
(54%).ȱCommonȱpracticeȱtoȱdrinkȱ3ȱorȱmoreȱenergyȱ
drinksȱwithȱalcoholȱ(49%).ȱWeeklyȱjoltȱandȱcrashȱ
episodesȱwereȱexperiencedȱbyȱ29%ȱofȱtheȱusers,ȱ
22%ȱreportedȱeverȱhavingȱheadaches,ȱandȱ19%ȱ
heartȱpalpitationsȱfromȱconsumingȱenergyȱdrinks.ȱ
697ȱstudentsȱ(24%)ȱreportedȱconsumingȱalcoholȱ O’Brienȱetȱal.ȱ
Unspecifiedȱenergyȱ Surveyȱwasȱconductedȱ
WebȬbasedȱ
2008ȱ
withȱenergyȱdrinks.ȱStudentsȱwhoȱwereȱmale,ȱ
drinksȱ
inȱ4271ȱcollegeȱstudentsȱ
randomȱ sampleȱ
white,ȱintramuralȱathletes,ȱfraternityȱorȱsororityȱ
fromȱ10ȱuniversitiesȱinȱ
surveyȱ
membersȱorȱpledges,ȱandȱyoungerȱwereȱ
NorthȱCarolinaȱ
significantlyȱmoreȱlikelyȱtoȱconsumeȱalcoholȱwithȱ
energyȱdrinks.ȱConsumptionȱofȱenergyȱdrinksȱwithȱ
alcoholȱwasȱassociatedȱwithȱsignificantlyȱincreasedȱ
heavyȱepisodicȱdrinking,ȱandȱtwiceȱasȱmanyȱ
episodesȱofȱweeklyȱdrunkenness.ȱStudentsȱ
consumingȱalcoholȱwithȱenergyȱdrinksȱalsoȱhadȱ
significantlyȱhigherȱprevalenceȱofȱalcoholȬrelatedȱ
consequences.ȱ
Oteriȱetȱal.ȱ
56.9%ȱstudentsȱdeclaredȱusingȱenergyȱdrinks.ȱ
Unspecifiedȱenergyȱ
450ȱcollegeȱstudentsȱ
2007ȱ
48.4%ȱofȱusersȱfrequentlyȱassociatedȱenergyȱdrinksȱ
drinksȱ
filledȱoutȱtheȱ
Questionnaireȱ
withȱalcohol.ȱ
questionnaireȱ
Thombsȱetȱal.ȱ
3Ȭfoldȱincreasedȱriskȱofȱleavingȱaȱbarȱhighlyȱ
Randomized,ȱ
Unspecifiedȱenergyȱ 802ȱpatronsȱfromȱ7ȱbarsȱ
2010ȱ
intoxicatedȱandȱ4Ȭfoldȱincreasedȱriskȱofȱintendingȱ
fieldȱinterviewȱ
drinksȱ
overȱ4ȱconsecutiveȱ
nights.ȱEveryȱ3rdȱpatronȱ toȱdriveȱuponȱleavingȱtheȱbarȱdistrict,ȱcomparedȱtoȱ
patronsȱwhoȱdidȱnotȱmixȱenergyȱdrinksȱwithȱ
exitingȱbarȱwasȱ
alcohol.ȱ
approached.ȱ
Tableȱ5.ȱStudiesȱinvestigatingȱtheȱenergyȱdrinksȱconsumedȱwithȱalcoholȱ
ȱ
ȱ
www.intechopen.com
ȱ
9
!
Subjectȱ
Productȱ
Effectsȱ
Outcome
6ȱdrinksȱ
Death;ȱclearȱ
Hemorrhagicȱ
Femaleȱ madeȱfromȱ
causeȱnotȱ
19ȱyȱ RedȱBullȱandȱ pulmonaryȱedemaȱ
establishedȱ
vodkaȱ
DrankȱRedȱ Lossȱofȱresponsiveness,ȱ Death;ȱclearȱ
Femaleȱ
causeȱnotȱ
collapse,ȱventricularȱ
Bullȱplusȱ
31ȱyȱ
establishedȱ
fibrillationȱ
vodkaȱ
Acuteȱseizures,ȱ
Unknown,ȱ
DrankȱRedȱ
epilepsyȱasȱchildȱbutȱ
Maleȱȱ20ȱ
clearȱcauseȱnotȱ
Bullȱplusȱ
noȱfurtherȱattacksȱsinceȱ
yȱ
establishedȱ
vodkaȱ
8ȱyȱ
Tableȱ6.ȱSwedishȱcaseȱreportsȱinvestigatingȱenergyȱdrinksȱmixedȱ
withȱalcoholicȱbeveragesȱ
Inȱadditionȱtoȱalcoholȱconsumption,ȱenergyȱdrinksȱmayȱbeȱ
usedȱ alongȱ withȱ prescriptionȱ medicationȱ andȱ recreationalȱ
drugs.ȱ Diamondȱ etȱ al.ȱ (2000)ȱ demonstratedȱ thatȱ caffeineȱ
canȱ enhanceȱ theȱ effectȱ ofȱ certainȱ analgesics,ȱ includingȱ
ibuprofen.ȱ Staibȱ etȱ al.ȱ (1987)ȱ advisedȱ thatȱ caffeineȱ intakeȱ
shouldȱ beȱ reducedȱ inȱ individualsȱ prescribedȱ certainȱ
antibioticsȱ asȱ theyȱ mayȱ inhibitȱ theȱ eliminationȱ ofȱ caffeineȱ
fromȱ theȱ body.ȱ Studiesȱ conductedȱ inȱ humansȱ showȱ thatȱ
caffeineȱ producesȱ subjectiveȱ andȱ behaviouralȱ effectsȱ thatȱ
areȱsimilarȱtoȱthoseȱofȱtypicalȱpsychomotorȱstimulantȱdrugsȱ
thatȱ areȱ knownȱ toȱ beȱ dopaminergicallyȱ mediatedȱ (e.g.,ȱ
amphetamines,ȱ cocaine).ȱ Caffeine,ȱ likeȱ amphetamineȱ andȱ
cocaine,ȱ enhancesȱ feelingsȱ ofȱ wellȬbeing,ȱ motivationȱ forȱ
work,ȱ energy,ȱ andȱ concentration,ȱ delaysȱ sleep,ȱ andȱ
enhancesȱ vigilanceȱ performanceȱ onȱ psychomotorȱ tasksȱ
(Garrettȱ&ȱGriffithsȱ1997).ȱȱ
ȱ
Overall,ȱ theȱ evidenceȱ doesȱ notȱ conclusivelyȱ indicateȱ aȱ
harmfulȱ toxicologicalȱ interactionȱ betweenȱ energyȱ drinksȱ
andȱ alcohol.ȱ However,ȱ theȱ dataȱ areȱ limitedȱ andȱ thereȱ isȱ
substantialȱ uncertainty.ȱ Itȱ wasȱ notedȱ thatȱ adverseȱ eventsȱ
associatedȱwithȱenergyȱdrinkȱandȱalcoholȱcoȬconsumptionȱ
happenedȱ mostȱ frequentlyȱ withȱ youngȱ adults.ȱ Thisȱ
affectedȱ subpopulationȱ isȱ relativelyȱ inexperiencedȱ asȱ
alcoholȱ consumersȱ andȱ relativelyȱ susceptibleȱ toȱ peerȱ
pressure.ȱ Theȱ adviceȱ toȱ recommendȱ notȱ toȱ mixȱ energyȱ
drinksȱ andȱ alcoholȱ couldȱ thereforeȱ beȱ arguedȱ toȱ beȱ aȱ
prudentȱsafetyȱmeasure.3ȱ
3.5ȱȱCanadianȱadverseȱreactionȱreportsȱ
Healthȱ Canada’sȱ Marketedȱ Healthȱ Productsȱ Directorateȱ
(MHPD)ȱ conductedȱ aȱ searchȱ ofȱ theȱ Canadaȱ Vigilanceȱ
AdverseȱReactionȱDatabase4ȱfromȱJanuaryȱ01,ȱ1965ȱtoȱJulyȱ
3ȱ Whileȱ thisȱ manuscriptȱ wasȱ inȱ preparation,ȱ theȱ UKȱ Committeeȱ
onȱ Toxicityȱ ofȱ Chemicalsȱ inȱ Food,ȱ Consumerȱ Productsȱ andȱ theȱ
Environmentȱpublishedȱaȱstatementȱonȱtheȱinteractionȱofȱcaffeineȱ
andȱ alcoholȱ (2012).ȱ Theȱ Committeeȱ concludedȱ thatȱ theȱ overallȱ
currentȱevidenceȱdoesȱnotȱsupportȱaȱharmfulȱinteractionȱbetweenȱ
caffeineȱ andȱ alcohol.ȱ However,ȱ itȱ statesȱ thatȱ thereȱ isȱ substantialȱ
uncertaintyȱinȱthisȱconclusion.ȱ
4ȱTheȱnumberȱofȱadverseȱreportsȱinȱtheȱCanadaȱVigilanceȱAdverseȱ
Reactionȱ Databaseȱ shouldȱ notȱ beȱ usedȱ asȱ aȱ basisȱ forȱ determiningȱ
theȱ incidenceȱ ofȱ aȱ reactionȱ orȱ forȱ estimatingȱ riskȱ ofȱ aȱ particularȱ
10 Int. food risk anal. j., 2013, Vol. 3, 4:2013
ȱ
20,ȱ 2010ȱ andȱ foundȱ 61ȱ adverseȱ reactionsȱ associatedȱ withȱ
theȱ consumptionȱ ofȱ energyȱ drinks.5ȱ ThirtyȬtwoȱ ofȱ theseȱ
reactionsȱ areȱ consideredȱ “serious”,ȱ 15ȱ ofȱ whichȱ involvedȱ
theȱ cardiacȱ systemȱ (arrhythmia,ȱ increasedȱ heartȱ rate,ȱ
palpitationsȱandȱchestȱpain).ȱSixȱofȱtheseȱ15ȱcardiacȱeventsȱ
occurredȱinȱindividualsȱagedȱ13Ȭ17ȱyears.ȱBasedȱonȱtheseȱ
reports,ȱ analysisȱ ofȱ availableȱ scientificȱ literature,ȱ MHPDȱ
detectedȱ aȱ safetyȱ signal6ȱ indicatingȱ anȱ associationȱ
betweenȱ adverseȱ cardiacȱ eventsȱ andȱ theȱ consumptionȱ ofȱ
energyȱ drinks.ȱ Almostȱ allȱ ofȱ theȱ reportsȱ areȱ inȱ healthyȱ
youngȱ individualsȱ withoutȱ concurrentȱ diseaseȱ orȱ
medications.ȱ Therefore,ȱ alternativeȱ causesȱ suchȱ asȱ
underlyingȱdiseaseȱorȱdrugȱinteractionȱwereȱnotȱapparent.ȱ
Fourȱ(4)ȱadverseȱcardiacȱreactionsȱwereȱassessedȱtoȱbeȱofȱ
probableȱ causalityȱ andȱ 8ȱ wereȱ assessedȱ toȱ beȱ ofȱ possibleȱ
causality,ȱ usingȱ theȱ WHOȱ causalityȱ assessmentȱ
algorithm6.ȱ Threeȱ adverseȱ reactions,ȱ includingȱ theȱ 2ȱ
deathsȱthatȱwereȱassociatedȱwithȱenergyȱdrinks,ȱcouldȱnotȱ
beȱassessedȱbecauseȱofȱlackȱofȱinformation.ȱTheȱincidenceȱ
ofȱcardiacȱeventsȱinȱtheȱhealthyȱyoungȱpopulationȱisȱrareȱ
andȱ notȱ wellȱ characterizedȱ inȱ theȱ scientificȱ literature.ȱ Inȱ
addition,ȱ theȱ absenceȱ ofȱ observationalȱ andȱ clinicalȱ trialȱ
dataȱonȱenergyȱdrinkȱuseȱpreventsȱaȱcomparisonȱbetweenȱ
reactionsȱ associatedȱ withȱ theseȱ productsȱ andȱ theȱ
backgroundȱincidenceȱofȱcardiacȱevents.ȱȱ
ȱ
Reportedȱ adverseȱ reactionsȱ ratesȱ areȱ knownȱ toȱ beȱ underȱ
reported,ȱ particularlyȱ forȱ NHPs.ȱ ȱ Bothȱ consumersȱ andȱ
healthcareȱ practitioners,ȱ generallyȱ regardȱ NHPsȱ asȱ safeȱ
andȱareȱusuallyȱunawareȱthatȱadverseȱreactionsȱcanȱoccurȱ
withȱ theseȱ productsȱ andȱ thatȱ theyȱ canȱ beȱ reportedȱ toȱ
Healthȱ Canada.ȱ Evenȱ thoughȱ energyȱ drinksȱ haveȱ beenȱ
regulatedȱ asȱ NHPsȱ (thereforeȱ healthȱ products,ȱ subjectȱ toȱ
adverseȱ reactionȱ reporting)ȱ fromȱ 2004ȱ toȱ 2011;ȱ theyȱ mayȱ
notȱ haveȱ beenȱ recognizedȱ asȱ such,ȱ givenȱ theȱ wayȱ theyȱ
haveȱ beenȱ madeȱ availableȱ (groceryȱ storesȱ nextȱ toȱ otherȱ
beverages,ȱ convenienceȱ stores,ȱ gasȱ stations).ȱ Beingȱ
perceivedȱ asȱ beveragesȱ (i.e.ȱ food)ȱ didȱ notȱ supportȱ
consistentȱ adverseȱ reactionȱ reportingȱ associatedȱ withȱ
theseȱproducts.ȱȱ
ȱ
Whileȱtheseȱadverseȱreactionsȱareȱimportantȱtoȱnote,ȱtheyȱareȱ
toȱ beȱ putȱ intoȱ perspective,ȱ consideringȱ theȱ comparativelyȱ
smallȱ numberȱ ofȱ reportedȱ adverseȱ reactionsȱ inȱ viewȱ ofȱ theȱ
numberȱ ofȱ productȱ unitsȱ soldȱ (aroundȱ 91ȱ millionȱ unitsȱ inȱ
Canadaȱ perȱ year,ȱ accordingȱ toȱ 2009ȱ ACNielsenȱ data).ȱ Thisȱ
comparisonȱisȱanȱoverȬsimplification.ȱItȱwouldȱbeȱbetterȱtoȱ
compareȱ theȱ numberȱ ofȱ adverseȱ effectsȱ toȱ theȱ numberȱ ofȱ
ȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱ ȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱ ȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱ
product,ȱasȱneitherȱtheȱtotalȱnumberȱofȱreactionsȱoccurring,ȱnorȱtheȱ
numberȱofȱpatientsȱexposedȱtoȱtheȱhealthȱproduct,ȱisȱknown.ȱ
5ȱTheȱfirstȱauthorizedȱenergyȱdrinkȱwasȱRedȱBullȱinȱ2004.ȱ
6ȱ Aȱ signalȱ isȱ reportedȱ informationȱ onȱ aȱ possibleȱ causalȱ
relationshipȱ betweenȱ anȱ adverseȱ eventȱ andȱ aȱ drug,ȱ theȱ
relationshipȱ beingȱ unknownȱ orȱ incompletelyȱ documentedȱ
previouslyȱ(WorldȱHealthȱOrganizationȱ1991).ȱ
www.intechopen.com
!
consumersȱ ofȱ energyȱ drinks,ȱ butȱ atȱ thisȱ time,ȱ Canadianȱ
consumptionȱdataȱareȱnotȱavailable.ȱNonetheless,ȱitȱcanȱbeȱ
saidȱ thatȱ theȱ numberȱ ofȱ adverseȱ reactionsȱ reportedȱ toȱ
Healthȱ Canadaȱ isȱ relativelyȱ smallȱ comparedȱ toȱ generalȱ
consumption.ȱȱ
3.6ȱȱSummaryȱȱ
Energyȱ drinks,ȱ likeȱ otherȱ beveragesȱ containingȱ addedȱ
sugars,ȱ areȱ oftenȱ associatedȱ withȱ aȱ highȱ caloricȱ intake.ȱ
Theseȱ drinksȱ areȱ notȱ toȱ beȱ consideredȱ “sportsȱ drinks”ȱ asȱ
theyȱ containȱ atȱ leastȱ oneȱ ingredientȱ thatȱ isȱ aȱ stimulantȱ
(caffeine).ȱ Despiteȱ someȱ studyȱ resultsȱ suggestingȱ thatȱ
energyȱ drinkȱ consumptionȱ priorȱ toȱ exerciseȱ mayȱ haveȱ
someȱperformanceȱbenefits,ȱthereȱareȱconcerns,ȱasȱcaffeineȱ
mayȱdelayȱtheȱreturnȱtoȱaȱrestingȱheartȱrate.ȱLimitedȱstudyȱ
resultsȱhaveȱshownȱincreasesȱinȱsystolicȱbloodȱpressureȱandȱ
heartȱ rateȱ associatedȱ withȱ moderateȱ intakesȱ ofȱ energyȱ
drinksȱ (2ȱ servings).ȱ However,ȱ acrossȱ theȱ variousȱ studies,ȱ
transitoryȱ changesȱ inȱ bloodȱ pressureȱ didȱ notȱ reachȱ
hypersensitiveȱ levelsȱ withȱ consumptionȱ ofȱ energyȱ drinks.ȱ
Theseȱ effectsȱ wereȱ deemedȱ toȱ beȱ similarȱ toȱ theȱ effectsȱ onȱ
bloodȱ pressureȱ demonstratedȱ inȱ conjunctionȱ withȱ caffeineȱ
intakeȱ(Riksenȱetȱal.,ȱ2009).ȱȱ
ȱ
Surveysȱalsoȱsuggestȱthatȱinȱsocialȱsettingsȱ(e.g.,ȱbars,ȱparties)ȱ
moderateȱ orȱ highȱ consumptionȱ ofȱ energyȱ drinksȱ withȱ
alcoholȱisȱnotȱuncommon,ȱandȱcouldȱleadȱtoȱriskyȱbehaviourȱ
sinceȱ energyȱ drinksȱ mayȱ maskȱ theȱ effectsȱ ofȱ alcohol.ȱ Thereȱ
haveȱ beenȱ adverseȱ reactionsȱ associatedȱ withȱ theȱ
consumptionȱ ofȱ energyȱ drinks,ȱ whetherȱ aloneȱ orȱ inȱ
combinationȱwithȱotherȱphysiologicallyȱactiveȱsubstances,ȱinȱ
Canada.ȱOnlyȱ4ȱofȱtheȱ15ȱreportedȱcardiacȱadverseȱreactionsȱ
wereȱ assessedȱ asȱ havingȱ aȱ probableȱ causality.ȱ Theȱ
significanceȱ ofȱ theseȱ reportedȱ adverseȱ reactionsȱ isȱ toȱ beȱ
discussedȱinȱtheȱcontextȱofȱtheirȱlimitedȱnumberȱversusȱtheȱ
numberȱofȱenergyȱdrinkȱunitsȱconsumedȱandȱtheȱabsenceȱofȱ
observationalȱ andȱ clinicalȱ trialȱ data,ȱ whichȱ preventsȱ aȱ
comparisonȱ betweenȱ reactionsȱ associatedȱ withȱ theseȱ
productsȱandȱtheȱ“background”ȱincidenceȱofȱcardiacȱevents.ȱȱ
ȱ
Thereȱwereȱinsufficientȱtoxicologicalȱdataȱtoȱcharacterizeȱtheȱ
healthȱ hazardȱ associatedȱ withȱ energyȱ drinksȱ asȱ aȱ singleȱ
product.ȱAsȱaȱconsequence,ȱtheȱmajorȱindividualȱingredientsȱ
wereȱ assessedȱ forȱ hazardȱ identificationȱ andȱ hazardȱ
characterization,ȱ andȱ thisȱ wasȱ consideredȱ aȱ meansȱ toȱ gainȱ
insightȱintoȱtheȱhazardȱcharacterizationȱofȱenergyȱdrinks.ȱȱ
4.ȱAssessmentȱofȱIndividualȱIngredientsȱofȱEnergyȱDrinksȱ
4.1ȱȱIntroductionȱ
Inȱotherȱjurisdictions,ȱtheȱapproachȱusedȱforȱassessmentsȱ
onȱ energyȱ drinksȱ consideredȱ theȱ nutritionalȱ andȱ
toxicologicalȱaspectsȱofȱtheȱindividualȱingredientsȱinȱtheseȱ
www.intechopen.com
ȱ
products.ȱThisȱapproachȱwasȱtakenȱbecauseȱtheȱpublishedȱ
literatureȱ onȱ energyȱ drinksȱ wasȱ limitedȱ toȱ aȱ fewȱ clinicalȱ
studiesȱandȱcaseȱreports.ȱNoȱrelevantȱpreclinicalȱ(animal)ȱ
studiesȱ haveȱ beenȱ conducted.ȱ Oneȱ studyȱ didȱ administerȱ
anȱenergyȱdrinkȱtoȱmiceȱforȱupȱtoȱ13ȱweeksȱbutȱtheȱworkȱ
wasȱnotȱconsideredȱtoȱbeȱusefulȱinȱtheȱsafetyȱassessment,ȱ
sinceȱ theȱ amountȱ consumedȱ didȱ notȱ constituteȱ aȱ veryȱ
highȱ doseȱ ofȱ theȱ energyȱ drinkȱ (citedȱ inȱ EFSAȱ 2009).ȱ Inȱ
contrast,ȱ thereȱ isȱ aȱ goodȱ biologicalȱ understandingȱ andȱ aȱ
broadȱ toxicologicalȱ databaseȱ forȱ manyȱ ofȱ theȱ individualȱ
ingredientsȱ thatȱ constituteȱ energyȱ drinks.ȱ Therefore,ȱ thisȱ
approachȱwasȱalsoȱusedȱinȱthisȱhazardȱcharacterization.ȱȱ
ȱ
Energyȱ drinksȱ canȱ beȱ generallyȱ characterisedȱ asȱ
containingȱ theȱ followingȱ ingredients:ȱ caffeine,ȱ taurine,ȱ
glucuronolactone,ȱ inositolȱ andȱ aȱ varietyȱ ofȱ Bȱ vitamins,ȱ
includingȱ thiamine,ȱ niacin,ȱ vitaminȱ B6,ȱ vitaminȱ B12,ȱ
pantothenicȱacidȱandȱriboflavinȱ(asȱmentionedȱinȱTableȱ1).ȱ
Theȱ dietaryȱ sources,ȱ nutritionalȱ aspectsȱ andȱ toxicityȱ ofȱ
theseȱingredientsȱareȱreviewedȱbelow.ȱȱ
4.2ȱȱCaffeineȱ
Caffeineȱ isȱ consumedȱ asȱ aȱ naturalȱ constituentȱ ofȱ coffee,ȱ
tea,ȱchocolateȱandȱotherȱnaturalȱsources,ȱsuchȱasȱguaranaȱ
andȱ yerbaȱ mate.ȱ Itȱ isȱ alsoȱ usedȱ asȱ aȱ foodȱ additiveȱ andȱ isȱ
presentȱinȱcertainȱcarbonatedȱsoftȱdrinks.ȱItȱisȱalsoȱpresentȱ
inȱsomeȱtherapeuticȱproducts,ȱsuchȱasȱcoldȱremediesȱandȱ
allergyȱ medicines.ȱ Inȱ Canada,ȱ itȱ isȱ estimatedȱ thatȱ maleȱ
andȱ femaleȱ adultsȱ haveȱ aȱ respectiveȱ meanȱ intakeȱ ofȱ 281ȱ
andȱ 230ȱ mgȱ ofȱ caffeineȱ perȱ dayȱ (Canadianȱ Communityȱ
Healthȱ Survey,ȱ 2004).ȱ Theseȱ amountsȱ areȱ aboutȱ threeȱ
timesȱ theȱ 80ȱ mgȱ ofȱ caffeineȱ presentȱ inȱ aȱ singleȱ 250ȱ mlȱ
servingȱofȱaȱtypicalȱenergyȱdrink.ȱȱ
ȱ
Whenȱ aȱ singleȱ servingȱ ofȱ aȱ caffeinatedȱ beverageȱ
containingȱ 40Ȭ120ȱ mgȱ ofȱ caffeine7ȱ isȱ consumedȱ theȱ mainȱ
biologicalȱ effectȱ ofȱ caffeineȱ isȱ thatȱ itȱ actsȱ asȱ aȱ stimulantȱ
andȱpromotesȱalertness,ȱenhancesȱcognitiveȱperformance,ȱ
relievesȱ fatigueȱ andȱ promotesȱ physicalȱ enduranceȱ
(DohertyȱandȱSmith,ȱ2004;ȱLoristȱandȱTops,ȱ2003;ȱAstorinoȱ
andȱ Robertson,ȱ 2010)ȱ Aȱ singleȱ servingȱ canȱ alsoȱ causeȱ
transientȱ adverseȱ effects,ȱ suchȱ asȱ insomnia,ȱ headachesȱ
andȱ nervousnessȱ inȱ caffeineȬsensitiveȱ individualsȱ
(Nawrotȱetȱal.ȱ2003;ȱHigdonȱandȱFrei,ȱ2006).ȱȱ
ȱ
Inȱ anȱ extensiveȱ reviewȱ ofȱ theȱ literature,ȱ Healthȱ Canadaȱ
assessedȱmoreȱthanȱ300ȱstudiesȱthatȱexaminedȱtheȱpotentialȱ
healthȱ effectsȱ ofȱ caffeineȱ (Nawrotȱ etȱ al.ȱ 2003).ȱ Itȱ wasȱ
concludedȱ thatȱ aȱ healthyȱ adultȱ couldȱ tolerateȱ aȱ maximumȱ
intakeȱofȱ400ȱmgȱofȱcaffeineȱperȱdayȱ(equivalentȱtoȱ6ȱmg/kgȱ
bw/dayȱforȱaȱ65ȱkgȱadult).ȱThisȱamountȱofȱcaffeineȱwasȱnotȱ
associatedȱ withȱ adverseȱ effectsȱ suchȱ asȱ generalȱ toxicity,ȱ
cardiovascularȱ effects,ȱ effectsȱ onȱ boneȱ status,ȱ changesȱ inȱ
ȱGenerally,ȱaȱservingȱofȱteaȱorȱcolaȱcanȱcontainȱaroundȱ40ȱmgȱofȱ
caffeineȱ whileȱ coffee,ȱ dependingȱ onȱ theȱ typeȱ andȱ brewingȱ
method,ȱcanȱcontainȱaroundȱ120ȱmgȱcaffeineȱandȱhigher.ȱ
7
11
!
behaviour,ȱ effectsȱ onȱ maleȱ fertilityȱ orȱ increasedȱ incidenceȱ
ofȱcancerȱdevelopment.ȱ
ȱ
Further,ȱ theȱ assessmentȱ concludedȱ thatȱ reproductiveȬ
agedȱ womenȱ couldȱ tolerateȱ aȱ maximumȱ intakeȱ ofȱ upȱ toȱ
300ȱmgȱcaffeineȱperȱdayȱ(equivalentȱtoȱ4.6ȱmg/kgȱbw/dayȱ
forȱ aȱ 65ȱ kgȱ adult),ȱ basedȱ onȱ reproductiveȱ considerationsȱ
(spontaneousȱ abortion,ȱ retardedȱ fetalȱ growth).ȱ Finallyȱ
children,ȱagedȱ4ȱtoȱ12ȱyears,ȱwereȱconsideredȱtoȱtolerateȱaȱ
maximumȱofȱ45ȱtoȱ85ȱmgȱofȱcaffeineȱperȱdayȱ(equivalentȱ
toȱ 2.5ȱ mg/kgȱ bw/dayȱ forȱ aȱ childȱ weighingȱ 18ȱ toȱ 34ȱ kg),ȱ
basedȱonȱtransientȱmildȱbehaviourȱchanges.ȱȱ
ȱ
Thereȱ wereȱ insufficientȱ dataȱ toȱ recommendȱ aȱ dailyȱ
maximumȱintakeȱofȱcaffeineȱforȱadolescents,ȱagedȱ13ȱtoȱ18ȱ
years.ȱ However,ȱ theȱ adultȱ recommendationȱ couldȱ beȱ
consideredȱ inappropriateȱ sinceȱ manyȱ adolescentsȱ mayȱ
haveȱaȱlighterȱbodyȱweightȱthanȱtheȱaverageȱadult.ȱMoreȱ
importantly,ȱ adolescentsȱ areȱ lessȱ developedȱ thanȱ adultsȱ
andȱ theirȱ growingȱ bodiesȱ mayȱ beȱ moreȱ susceptibleȱ toȱ
adverseȱeffectsȱofȱcaffeine.ȱOneȱauthorȱhasȱsuggestedȱthatȱ
caffeineȱ mayȱ adverselyȱ affectȱ theȱ growingȱ adolescentȱ
brainȱ (Templeȱ 2009).ȱ Alternatively,ȱ adolescentsȱ mayȱ beȱ
lessȱhabituatedȱtoȱconsumingȱcaffeineȱandȱthereforeȱmayȱ
beȱ moreȱ susceptible.ȱ Inȱ anyȱ case,ȱ thereȱ isȱ substantialȱ
uncertaintyȱ asȱ toȱ whetherȱ theȱ adultȱ recommendationȱ
shouldȱbeȱapplied.ȱAsȱaȱprecaution,ȱitȱwouldȱbeȱprudentȱ
toȱ ȱ recommendȱ thatȱ anȱ adolescentȱ haveȱ aȱdailyȱ intakeȱ ofȱ
caffeineȱ noȱ greaterȱ thanȱ theȱ amountȱ calculatedȱ fromȱ theȱ
doseȱ usedȱ toȱ determineȱ theȱ recommendedȱ maximalȱ
intakeȱforȱchildrenȱ(2.5ȱmg/kgȱbw/day)ȱandȱtheȱindividualȱ
adolescentȱ bodyȱ weightȱ (estimatedȱ rangeȱ ofȱ adolescentȱ
bodyȱ weights:ȱ 40Ȭ70ȱ kg).ȱ Thisȱ doseȱ wouldȱ suggestȱ thatȱ
adolescentsȱ couldȱ consumeȱ 100ȱ toȱ 175ȱ mgȱ ofȱ caffeineȱ
daily,ȱ dependingȱ onȱ theȱ individualȱ bodyȱ weightȱ ofȱ theȱ
adolescent.ȱȱ
ȱ
Itȱ shouldȱ beȱ notedȱ thatȱ allȱ ofȱ theseȱ recommendedȱ
maximumȱ intakesȱ ofȱ caffeineȱ areȱ consideredȱ aȱ dailyȱ
amountȱ thatȱ wouldȱ beȱ theȱ resultȱ ofȱ cumulativeȱ
consumptionȱ throughoutȱ theȱ day.ȱ Theȱ ingestionȱ ofȱ theȱ
dailyȱ maximumȱ intakeȱ inȱ aȱ briefȱ periodȱ ofȱ time,ȱ thatȱ isȱ
1Ȭ2ȱ hours,ȱ mayȱ causeȱ adverseȱ reactions,ȱ suchȱ asȱ
insomnia,ȱheadache,ȱstomachȱache,ȱnervousness,ȱnauseaȱ
orȱ moreȱ seriousȱ reactions.ȱ Forȱ example,ȱ studiesȱ haveȱ
shownȱ thatȱ aȱ singleȱ ingestionȱ ofȱ moreȱ thanȱ 250ȱ mgȱ ofȱ
caffeineȱbyȱaȱhealthyȱadultȱcanȱalsoȱcauseȱanȱincreaseȱinȱ
bloodȱpressureȱ(MaughanȱandȱGriffin,ȱ2003)ȱwhileȱmoreȱ
thanȱ 450ȱ mgȱ ofȱ caffeineȱ mayȱ resultȱ inȱ tachycardiaȱ
(Nawrotȱetȱal.ȱ2003).ȱȱ
4.3ȱȱTaurineȱ
Taurineȱ isȱ anȱ aminoȱ acidȱ thatȱ isȱ naturallyȱ presentȱ inȱ theȱ
diet.ȱ Specifically,ȱ itȱ isȱ foundȱ inȱ meatȱ andȱ seafood.ȱ
Estimatesȱofȱhumanȱdailyȱintakeȱofȱtaurineȱrangeȱfromȱ40ȱ
mgȱtoȱ400ȱmgȱ(HayesȱandȱTrautweinȱ1994).ȱTheseȱdietaryȱ
ȱ
levelsȱ ofȱ taurineȱ areȱ relativelyȱ lowȱ comparedȱ withȱ theȱ
1000ȱmgȱofȱtaurineȱpresentȱinȱaȱsingleȱservingȱofȱaȱtypicalȱ
energyȱdrink.ȱȱ
ȱ
Taurineȱ isȱ alsoȱ synthesizedȱ inȱ theȱ liverȱ fromȱ theȱ aminoȱ
acidȱcysteine,ȱasȱwellȱasȱfromȱotherȱsulphurȱcompounds.ȱ
Itȱhasȱaȱroleȱinȱseveralȱbiologicalȱprocesses,ȱincludingȱtheȱ
formationȱ ofȱ bileȱ salt,ȱ cellȱ membraneȱ stability,ȱ theȱ
modulationȱ ofȱ calciumȱ flowȱ andȱ neuronalȱ excitabilityȱ
(FSPBȱ 2002).ȱ Taurineȱ isȱ consideredȱ essentialȱ forȱ theȱ
normalȱ developmentȱ ofȱ infantsȱ andȱ consequentlyȱ itȱ isȱ aȱ
standardȱ ingredientȱ inȱ infantȱ formula,ȱ suchȱ thatȱ
newbornsȱingestȱaboutȱ280ȱmgȱdailyȱ(equivalentȱtoȱaȱdoseȱ
ofȱ17ȱmg/kgȱbw/day).ȱȱ
ȱ
Taurineȱ isȱ oneȱ ofȱ theȱ mostȱ abundantȱ aminoȱ acidsȱ inȱ theȱ
humanȱ body.ȱ Itȱ isȱ presentȱ inȱ relativelyȱ highȱ amountsȱ inȱ
skeletalȱ andȱ cardiacȱ muscleȱ (Timbrellȱ etȱ al.,ȱ 1995)ȱ andȱ
evidenceȱ inȱ experimentalȱ animalsȱ suggestsȱ thatȱ itȱ isȱ
essentialȱ toȱ normalȱ muscleȱ maintenanceȱ andȱ functionȱ
(Warskulatȱ etȱ al.,ȱ 2007).ȱ Limitedȱ literatureȱ hasȱ suggestedȱ
thatȱ taurineȱ supplementsȱ mayȱ haveȱ aȱ beneficialȱ effectȱ inȱ
personsȱ withȱ congestiveȱ heartȱ failure,ȱ hypertension,ȱ
diabetesȱandȱskeletalȱmuscleȱdisordersȱ(referencesȱcitedȱinȱ
Bouknoogheȱetȱal.,ȱ2006;ȱShaoȱandȱHathcock,ȱ2008).ȱȱ
ȱ
Humanȱ studiesȱ suggestȱ thatȱ taurineȱ isȱ readilyȱ absorbedȱ
fromȱ consumedȱ foodȱ andȱ thatȱ plasmaȱ levelȱ ofȱ taurineȱ
peakȱ withinȱ anȱ hourȱ afterȱ ingestionȱ (SCFȱ 2003).ȱ Thisȱ
observationȱ isȱ consistentȱ withȱ theȱ findingsȱ ofȱ aȱ studyȱ
whereȱ ratsȱ receivedȱ taurineȱ asȱ aȱ singleȱ oralȱ doseȱ ofȱ 300ȱ
mg/kgȱbw.ȱItȱwasȱobservedȱthatȱtheȱexogenousȱtaurineȱisȱ
quicklyȱ absorbedȱ intoȱ theȱ body,ȱ equilibratesȱ withȱ
endogenousȱpoolsȱofȱtaurineȱandȱthatȱexcessȱamountsȱareȱ
rapidlyȱeliminatedȱbyȱtheȱkidneys.ȱTheȱstudyȱalsoȱshowedȱ
thatȱ aȱ 14Ȭdayȱ repeatȱ treatmentȱ withȱ taurineȱ didȱ notȱ
changeȱ theȱ fateȱ ofȱ theȱ lastȱ singleȱ oralȱ doseȱ (Svedȱ etȱ al.ȱ
2007),ȱ whichȱ suggestsȱ thatȱ theȱ bodyȱ canȱ metabolicallyȱ
handleȱrelativelyȱlargeȱamountsȱofȱtaurineȱdaily.ȱ
ȱ
Theȱ acuteȱ oralȱtoxicityȱ ofȱ taurineȱ isȱconsideredȱ relativelyȱ
low,ȱ suchȱ thatȱ noȱ adverseȱ effectsȱ haveȱ beenȱ observedȱ
followingȱaȱsingleȱadministrationȱinȱratsȱupȱtoȱ7000ȱmg/kgȱ
bwȱorȱinȱhumansȱupȱtoȱ150ȱmg/kgȱbwȱ(equalȱtoȱ10,500ȱmgȱ
forȱaȱ70ȱkgȱadult).ȱȱ
ȱ
Inȱ shortȬtermȱ studiesȱ withȱ humanȱ subjects,ȱ theȱ ingestionȱ
ofȱ upȱ toȱ 6000ȱ mgȱ perȱ dayȱ forȱ 42ȱ daysȱ andȱ 1500ȱ mgȱ perȱ
dayȱ forȱ 90ȱ daysȱ showedȱ noȱ evidenceȱ ofȱ adverseȱ effectsȱ
(SCFȱ 2003).ȱ Inȱ aȱ 90Ȭdayȱ studyȱ inȱ rats,ȱ taurineȱ wasȱ
administeredȱ byȱ gavageȱ toȱ groupsȱ ofȱ animalsȱ (20ȱ
animals/sex/dose)ȱ atȱ dosesȱ ofȱ 0,ȱ 300,ȱ 600ȱ orȱ 1000ȱ mg/kgȱ
bw/day.ȱ Noȱ taurineȬrelatedȱ changesȱ ofȱ standardȱ
toxicologicalȱ parametersȱ wereȱ observed.ȱ Theȱ exceptionȱ
wasȱ aȱ doseȬrelatedȱ behaviouralȱ changeȱ (increasedȱ
activity,ȱselfȬinjury)ȱthatȱwasȱobservedȱinȱallȱthreeȱtreatedȱ
www.intechopen.com
!
groupsȱ andȱ inȱ bothȱ sexesȱ ofȱ theȱ testȱ animalȱ (SCFȱ 2003).ȱ
Subsequently,ȱ aȱ secondȱ 90Ȭdayȱ studyȱ wasȱ conducted,ȱ
whereȱ taurineȱ wasȱ administeredȱ byȱ gavageȱ toȱ groupsȱ ofȱ
ratsȱ(20ȱanimals/sex/dose)ȱatȱdosesȱofȱ0,ȱ600ȱorȱ1000ȱmg/kgȱ
bw/day,ȱ andȱ anotherȱ setȱ ofȱ ratsȱ (20ȱ animals/sex/dose)ȱ
receivedȱ taurineȱ inȱ drinkingȱ waterȱ atȱ dosesȱ ofȱ 0,ȱ 1000ȱ orȱ
1500ȱmg/kgȱbw/day.ȱInȱadditionȱtoȱstandardȱtoxicologicalȱ
parameters,ȱ aȱ functionalȱ observationalȱ batteryȱ andȱ
locomotorȱ activityȱ dataȱ wereȱ collectedȱ atȱ 0,ȱ 6ȱ andȱ 12ȱ
weeksȱofȱtheȱstudyȱandȱconductedȱinȱaȱblindȱmannerȱ(theȱ
previousȱ 90Ȭdayȱ study’sȱ findingsȱ wereȱ arguedȱ byȱ theȱ
study’sȱauthorȱtoȱbeȱbiasedȱsinceȱtheyȱwereȱnotȱblinded).ȱ
Theȱ resultsȱ indicatedȱ thatȱ thereȱ wereȱ noȱ taurineȬrelatedȱ
deaths,ȱ clinicalȱ toxicities,ȱ bodyȱ weightȱ changes,ȱ foodȱ orȱ
waterȱ consumptionȱ differences,ȱ orȱ macroscopicȱ changes.ȱ
Theȱ functionalȱ observationalȱ batteryȱ andȱ locomotorȱ
activityȱ dataȱ didȱ notȱ showȱ anyȱ taurineȬrelatedȱ effect.ȱ
Thereȱ wereȱ noȱ adverseȱ behaviouralȱ effectsȱ observedȱ inȱ
thisȱsecondȱstudy.ȱAȱNOAELȱforȱtaurineȱwasȱestablishedȱ
atȱ theȱ levelȱ ofȱ 1000ȱ mg/kgȱ bw/day,ȱ basedȱ onȱ theȱ highestȱ
doseȱ testedȱ inȱ theȱ firstȱ 90Ȭdayȱ study,ȱ whichȱ includedȱ aȱ
histopathologicalȱassessmentȱ(EFSAȱ2009).ȱȱ
ȱ
Aȱ developmentalȱ toxicityȱ studyȱ inȱ miceȱ showedȱ thatȱ
whenȱ orallyȱ administeredȱ atȱ aȱ doseȱ ofȱ 4000ȱ mg/kgȱ
bw/dayȱ onȱ daysȱ 7ȱ toȱ 14ȱ ofȱ gestation,ȱ taurineȱ wasȱ notȱ aȱ
teratogen,ȱthatȱis,ȱitȱdidȱnotȱcauseȱbirthȱdefectsȱ(Takahashiȱ
etȱal.ȱ1972).ȱȱ
ȱ
Inȱ theȱ additionalȱ studiesȱ conducted,ȱ taurineȱ hasȱ notȱ
shownȱ anyȱ mutagenicȱ orȱ genotoxicȱ potentialȱ inȱ bacterialȱ
orȱmammalianȱcellȱinȱvitroȱassayȱsystemsȱ(S.A.ȱLaidlawȱetȱ
al.ȱ1989;ȱR.ȱCossiȱetȱal.ȱ1995).ȱ
ȱ
ThereȱareȱnoȱlongȬtermȱtoxicityȱorȱcarcinogenicityȱstudiesȱ
conductedȱtoȱassessȱtheȱcarcinogenicȱpotentialȱofȱtaurine;ȱ
however,ȱ thereȱ isȱ noȱ indicationȱ thatȱ taurineȱ isȱ aȱ
carcinogenȱbasedȱonȱshortȬtermȱtoxicityȱstudies.ȱ
ȱ
Inȱoverȱthirtyȱstudiesȱwithȱadult,ȱchildȱandȱinfantȱsubjects,ȱ
theȱ useȱ ofȱ taurineȱ hasȱ notȱ demonstratedȱ anyȱ safetyȱ
concerns.ȱMostȱofȱtheseȱstudiesȱinvolvedȱtheȱingestionȱofȱ
taurineȱonȱaȱdailyȱbasisȱwithȱdosesȱinȱtheȱrangeȱofȱ3000ȱtoȱ
6000ȱmgȱforȱperiodsȱofȱupȱtoȱoneȱmonthȱorȱlongerȱwithoutȱ
anyȱapparentȱadverseȱhealthȱeffectsȱ(EFSAȱ2009).ȱȱ
ȱ
Oneȱ doubleȬblindȱ studyȱ inȱ humanȱ volunteersȱ (14ȱ
subjects)ȱ showedȱ thatȱ aȱ combinationȱ ofȱ caffeineȱ andȱ
taurine,ȱatȱlevelsȱpresentȱinȱaȱtypicalȱenergyȱdrink,ȱhadȱnoȱ
effectȱ onȱ shortȬtermȱ memory,ȱ asȱ indicatedȱ byȱ anȱ
abbreviatedȱ versionȱ ofȱ aȱ standardȱ testȱ forȱ shortȬtermȱ
memory.ȱ However,ȱ thisȱ combinationȱ didȱ induceȱ aȱ
decreaseȱ inȱ heartȱ rateȱ andȱ anȱ increaseȱ inȱ meanȱ arterialȱ
bloodȱpressure.ȱThisȱfindingȱisȱunexpectedȱsinceȱcaffeineȱ
generallyȱstimulatesȱheartȱrate.ȱFurtherȱinvestigationȱintoȱ
theȱ combinationȱ ofȱ theseȱ twoȱ substancesȱ isȱ warrantedȱ
(Bichlerȱetȱal.,ȱ2006).ȱ
www.intechopen.com
ȱ
4.4ȱȱDȬGlucuronoȬ·Ȭlactoneȱ
DȬGlucuronoȬ·Ȭlactoneȱ (glucuronolactone)ȱ isȱ theȱ ·Ȭ
lactoneȱ ofȱ glucuronicȱ acid.ȱ Itȱ isȱ aȱ normalȱ humanȱ
metaboliteȱandȱformedȱfromȱglucoseȱandȱglucuronicȱacid.ȱ
Itȱseemsȱtoȱbeȱfoundȱnaturallyȱinȱonlyȱaȱsmallȱnumberȱofȱ
foodsȱsuchȱasȱwineȱandȱplantȱgumsȱ(e.g.ȱguarȱgum,ȱgumȱ
Arabic).ȱ Anȱ estimateȱ ofȱ theȱ meanȱ dailyȱ intakeȱ isȱ 1.2ȱ
mg/dayȱ andȱ aȱ highȱ dailyȱ intakeȱ isȱ suggestedȱ toȱ beȱ 2.3ȱ
mg/dayȱ (SCFȱ 1999).ȱ Theseȱ intakeȱ valuesȱ areȱ veryȱ smallȱ
whenȱcomparedȱtoȱtheȱintakeȱofȱglucuronolactoneȱofȱ600ȱ
mgȱfromȱtheȱconsumptionȱofȱaȱsingleȱservingȱofȱaȱtypicalȱ
energyȱ drink.ȱ Itȱ isȱ claimedȱ thatȱ glucuronolactoneȱ isȱ aȱ
quickȱ energyȱ sourceȱ andȱ mayȱ assistȱ inȱ theȱ detoxificationȱ
ofȱxenobioticsȱ(SCFȱ1999).ȱ
ȱ
Humanȱ andȱ ratȱ studiesȱ showȱ thatȱ whenȱ ingested,ȱ
glucuronolactoneȱ isȱ rapidlyȱ absorbed,ȱ metabolisedȱ andȱ
excretedȱ asȱ glucaricȱ acid,ȱ xylitolȱ andȱ LȬxylulose.ȱ Theseȱ
compoundsȱ areȱ notȱ consideredȱ toȱ beȱ toxicologicallyȱ
significantȱ (ANZFAȱ 2001).ȱ Inȱ contrastȱ toȱ humans,ȱ miceȱ
andȱ ratsȱ haveȱ anȱ additionalȱ metabolicȱ pathwayȱ thatȱ
allowsȱthemȱtoȱuseȱglucuronicȱacidȱtoȱsynthesizeȱvitaminȱ
C.ȱ Rodentsȱ canȱ alsoȱ useȱ exogenousȱ glucuronolactoneȱ toȱ
yieldȱ glucuronicȱ acidȱ andȱ thenȱ generateȱ vitaminȱ C.ȱ Thisȱ
additionalȱ pathwayȱ inȱ theȱ rodentȱ createdȱ someȱ
uncertaintyȱwithȱrespectȱtoȱtheȱappropriatenessȱofȱrodentsȱ
asȱ aȱ modelȱ forȱ humansȱ (SCFȱ 1999).ȱ However,ȱ furtherȱ
examinationȱ ofȱ theȱ issueȱ hasȱ determinedȱ thatȱ theȱ rodentȱ
metabolicȱ pathwayȱ isȱ relativelyȱ minorȱ inȱ theȱ animal’sȱ
handlingȱ ofȱ glucuronolactone,ȱ whichȱ suggestsȱ thatȱ
toxicologicalȱ resultsȱ fromȱ rodentsȱ areȱ relevantȱ toȱ theȱ
humanȱsituationȱ(SCFȱ2003.ȱ
ȱ
Theȱ acuteȱ oralȱ toxicityȱ ofȱ glucuronolactoneȱ isȱ veryȱ low,ȱ
suchȱ thatȱ inȱ miceȱ theȱ oralȱ LD50ȱ isȱ greaterȱ thanȱ 20000ȱ
mg/kgȱ bwȱ andȱ inȱ ratsȱ itȱ isȱ approximatelyȱ 11000ȱ mg/kgȱ
bw.ȱTheȱshortȬtermȱoralȱtoxicityȱofȱglucuronolactoneȱwasȱ
assessedȱ inȱ aȱ studyȱ whereȱ groupsȱ ofȱ ratsȱ (20ȱ
animals/sex/dose)ȱ wereȱ administeredȱ aȱ singleȱ dailyȱ
gavageȱdoseȱofȱ0,ȱ300,ȱ600ȱorȱ1000ȱmg/kgȱbw,ȱforȱupȱtoȱ90ȱ
daysȱ(SCFȱ2003).ȱTheȱresultsȱshowedȱnoȱtreatmentȬrelatedȱ
deaths,ȱ noȱ significantȱ differenceȱ inȱ bodyȱ weights,ȱ foodȱ
consumption,ȱ hematologicalȱ orȱ clinicalȱ chemistryȱ
parameters.ȱ Urinalysisȱ showedȱ thatȱ malesȱ treatedȱ withȱ
1000ȱmg/kgȱbwȱgroupȱhadȱurineȱwithȱaȱlowerȱpHȱthanȱtheȱ
controlȱ group,ȱ andȱ malesȱ inȱ theȱ 600ȱ andȱ 1000ȱ mg/kgȱ
groupsȱ hadȱ aȱ lowerȱ specificȱ gravityȱ thanȱ theȱ controlȱ
group.ȱ Resultsȱ fromȱ histopathologicalȱ examinationȱ
showedȱ vacuolisationȱ andȱ inflammatoryȱ changesȱ
localisedȱtoȱtheȱpapillaȱofȱtheȱkidneyȱinȱfemalesȱsuchȱthatȱ
atȱ dosesȱ ofȱ 0,ȱ 300,ȱ 600ȱ andȱ 1000ȱ mg/kgȱ bw/day,ȱ theȱ
incidencesȱwereȱrespectivelyȱ11/20,ȱ9/20,ȱ11/20ȱandȱ11/20.ȱ
TheȱincidenceȱofȱeffectȱwasȱnotȱdoseȬrelatedȱbutȱreviewersȱ
ofȱ theȱ studyȱ notedȱ thatȱ theȱseverityȱ ofȱ thisȱkidneyȱ lesionȱ
showedȱanȱincreaseȱwithȱdose.ȱAtȱdosesȱofȱ0,ȱ300,ȱ600ȱandȱ
1000ȱ mg/kgȱ bw/day,ȱ theȱ incidencesȱ ofȱ aȱ gradeȱ 2ȱ lesionȱ
13
!
(mildȱseverity)ȱwereȱrespectivelyȱ1/20,ȱ1/20,ȱ5/20ȱandȱ8/20.ȱ
Theȱ reviewersȱ concludedȱ thatȱ theȱ NOAELȱ forȱ theȱ studyȱ
wasȱ300ȱmg/kgȱbw/day.ȱȱ
ȱ
Inȱ aȱ secondȱ 90Ȭdayȱ studyȱ inȱ rats,ȱ theȱ toxicityȱ ofȱ
glucuronolactoneȱ wasȱ assessedȱ withȱ specificȱ focusȱ onȱ theȱ
kidneysȱ (SCFȱ 2009).ȱ Theȱ previousȱ gavageȱ studyȱ wasȱ
repeatedȱ withȱ anȱ additionalȱ fourȱ setsȱ ofȱ animalsȱ (20ȱ
animals/sex/dose)ȱ thatȱ receivedȱ glucuronolactoneȱ inȱ
drinkingȱwater;ȱwithȱbothȱroutesȱofȱadministrationȱanimalsȱ
receivedȱnominalȱdosesȱofȱȱ0,ȱ300,ȱ600ȱorȱ1000ȱmg/kgȱbwȱforȱ
90ȱdays.ȱThereȱwereȱnoȱtreatmentȬrelatedȱdeaths,ȱnoȱeffectsȱ
onȱclinicalȱobservations,ȱfoodȱorȱwaterȱconsumption,ȱbodyȱ
weights,ȱ clinicalȱ parameters,ȱ organȱ weightsȱ orȱ clinicalȱ
chemistryȱ parametersȱ relatedȱ toȱ renalȱ function.ȱ Urinalysisȱ
demonstratedȱ noȱ treatmentȬrelatedȱ effectsȱ andȱ noȱ
differencesȱ betweenȱ theȱ gavageȱ andȱ drinkingȱ waterȱ
groups.ȱLastlyȱthereȱwereȱnoȱtestȱmaterialȬrelatedȱgrossȱorȱ
microscopicȱ findings.ȱ Thisȱ includedȱ theȱ kidneysȱ whichȱ
showedȱ typicalȱ amountsȱ ofȱ backgroundȱ lesionsȱ forȱ thisȱ
strainȱ ofȱ rat.ȱ Thereȱ wasȱ noȱ testȱ materialȬrelatedȱ
vacuolisationȱofȱtheȱcellsȱliningȱtheȱcollectingȱtubulesȱofȱtheȱ
kidney.ȱ Thisȱ suggestsȱ thatȱ theȱ kidneyȱ lesionsȱ observedȱ inȱ
theȱ firstȱ studyȱ wereȱ notȱ significant,ȱ sinceȱ theȱ drinkingȱ
waterȱadministeredȱinȱtheȱsecondȱstudyȱwasȱmoreȱrelevantȱ
toȱtheȱhumanȱsituation.ȱTheȱNOAELȱforȱtheȱstudyȱwasȱsetȱ
atȱ1000ȱmg/kgȱbw/day,ȱtheȱhighestȱdoseȱtested.ȱ
ȱ
Reproductiveȱ andȱ developmentalȱ toxicityȱ studiesȱ forȱ
glucuronolactoneȱ wereȱ notȱ availableȱ butȱ wereȱ notȱ
consideredȱ necessaryȱ toȱ conduct,ȱ asȱ partȱ ofȱ theȱ safetyȱ
evaluationȱsinceȱglucuronolactoneȱinȱtheȱbodyȱhydrolysesȱ
toȱglucuronicȱacid,ȱwhichȱisȱanȱendogenousȱmetaboliteȱinȱ
humansȱandȱpresentȱinȱnormalȱhumanȱdietsȱ(EFSAȱ2009).ȱ
ȱ
Glucuronolactoneȱ wasȱ shownȱ notȱ toȱ beȱ mutagenicȱ inȱ aȱ
bacterialȱreverseȱmutationȱsystemȱ(Kurodaȱetȱal.ȱ1986).ȱȱ
ȱ
Oneȱ studyȱ (Ahrensȱ etȱ al.ȱ 1987)ȱ assessedȱ theȱ longȬtermȱ
toxicityȱ ofȱ glucuronolactone.ȱ Groupsȱ ofȱ 25ȱ maleȱ ratsȱ wereȱ
administeredȱorallyȱ0ȱmgȱ(water),ȱ125ȱmgȱofȱglucuronicȱacid,ȱ
250ȱmgȱofȱglucuronicȱacidȱorȱ125ȱmgȱofȱglucuronolactoneȱperȱ
dayȱfromȱtheȱageȱofȱaboutȱ1ȱyearȱtoȱdeathȱ(aboutȱ3ȱyearsȱofȱ
age).ȱTheȱdailyȱamountȱofȱglucuronolactoneȱwasȱequivalentȱ
toȱ aboutȱ 250ȱ mg/kgȱ bw/day.ȱ Thereȱ wereȱ noȱ significantȱ
differencesȱ betweenȱ theȱ treatmentȱ groupsȱ withȱ respectȱ
toȱ causeȱ ofȱ death,ȱ bodyȱ weightsȱ atȱ deathȱ orȱ autopsyȱ
findings.ȱ Theȱ carcinogenicȱ potentialȱ ofȱ glucuronolactoneȱ
wasȱ notȱ assessedȱ inȱ thisȱ orȱ anyȱ otherȱ study,ȱ however,ȱ
thereȱ isȱ noȱ indicationȱ thatȱ glucuronolactoneȱ isȱ aȱ
carcinogen,ȱ basedȱ onȱ shortȬtermȱ toxicityȱ studiesȱ andȱ itsȱ
roleȱinȱnormalȱhumanȱmetabolism.ȱȱ
ȱ
Itȱhasȱbeenȱreportedȱthatȱglucuronolactoneȱhasȱbeenȱusedȱ
atȱ dosesȱ ofȱ 1ȱ toȱ 3ȱ g/dayȱ (1000ȱ toȱ 3000ȱ mg/day)ȱ inȱ longȬ
termȱtherapyȱforȱcarriersȱofȱtheȱtyphoidȱorganismȱbecauseȱ
ofȱitsȱabilityȱtoȱinhibitȱviralȱandȱbacterialȱΆȬglucuronidase.ȱ
ȱ
Itsȱ useȱ wasȱ notȱ observedȱ toȱ causeȱ anyȱ adverseȱ effectsȱ
(KohlerȱandȱSchmidȱ1980).ȱ
4.5ȱȱInositolȱ
Myoinositolȱ
(inositol)ȱ
isȱ
aȱ
constituentȱ
ofȱ
phosphatidylinositol,ȱ aȱ phospholipid,ȱ whichȱ playsȱ anȱ
essentialȱroleȱinȱgrowth,ȱmetabolismȱregulationȱandȱsignalȱ
transduction.ȱ Inositolȱ isȱ aȱ normalȱ componentȱ ofȱ humanȱ
tissueȱandȱcanȱbeȱsynthesizedȱinȱsomeȱtissues.ȱItȱisȱaȱnormalȱ
partȱofȱfoodȱderivedȱfromȱplantsȱinȱtheȱformȱofȱphytateȱandȱ
fromȱ animalsȱ inȱ theȱ formȱ ofȱ freeȱ andȱ phosphorylatedȱ
inositolȱ andȱ asȱ inositolȱ phospholipid.ȱ Itȱ isȱ estimatedȱ thatȱ
adultsȱ ingestȱ aboutȱ 500ȱ toȱ 1000ȱ mgȱ ofȱ inositolȱ daily.ȱ Thisȱ
amountȱ isȱ relativelyȱ largeȱ comparedȱ toȱ theȱ 50ȱ mgȱ ofȱ
inositolȱ presentȱ inȱ aȱ singleȱ servingȱ ofȱ aȱ typicalȱ energyȱ
drink,ȱasȱdescribedȱinȱTableȱ1.ȱPotentialȱbenefitsȱofȱinositolȱ
mayȱ includeȱ decreasedȱ cholesterolȱ levelsȱ andȱ thusȱ aȱ
loweredȱriskȱofȱcardiovascularȱdiseaseȱ(ANZFAȱ2001).ȱȱ
ȱ
Theȱ toxicityȱ associatedȱ withȱ inositolȱ isȱ veryȱ low.ȱ Inȱ miceȱ
theȱoralȱLD50ȱisȱreportedȱtoȱbeȱ10000ȱmg/kgȱbw.ȱThereȱareȱ
noȱ reproductiveȱ orȱ developmentalȱ toxicityȱ studiesȱ orȱ
genotoxicityȱ studiesȱ thatȱ assessedȱ inositol.ȱ Althoughȱ
inositolȱ wasȱ notȱ testedȱ asȱ aȱ carcinogen,ȱ severalȱ studiesȱ
assessedȱ itsȱ abilityȱ toȱ preventȱ cancerȱ developmentȱ inȱ
mouseȱ models.ȱ Theȱ resultsȱ showedȱ thatȱ aȱ 3%ȱ levelȱ inȱ theȱ
dietȱ (equivalentȱ toȱ aȱ doseȱ ofȱ 6000ȱ mg/kgȱ bw/day)ȱ didȱ notȱ
increaseȱcancerȱformationȱ(EstensenȱandȱWattenbergȱ1993).ȱȱ
ȱ
Inȱ humans,ȱ inositolȱ hasȱ beenȱ usedȱ asȱ anȱ experimentalȱ
therapyȱ forȱ depression,ȱ panicȱ disorder,ȱ andȱ obsessiveȱ
compulsiveȱdisorder.ȱThereȱisȱaȱreportȱofȱpeopleȱconsumingȱ
upȱtoȱ20,000ȱmgȱofȱinositolȱdailyȱforȱ2ȱweeksȱ(Arendrupȱetȱal.ȱ
1989).ȱ Anotherȱ reportȱ citesȱ theȱ administrationȱ ofȱ 18ȱ gȱ ofȱ
inositolȱ dailyȱ forȱ 6ȱ weeksȱ (Koponenȱ etȱ al.ȱ 1997).ȱ Inȱ theseȱ
reportsȱandȱseveralȱothers,ȱnoȱadverseȱeffectsȱwereȱobservedȱ
(ColondyȱandȱHoffmanȱ1998).ȱȱ
4.6ȱȱBȱVitaminsȱ
Mostȱ energyȱ drinksȱ containȱ addedȱ Bȱ vitaminsȱ includingȱ
thiamine,ȱriboflavin,ȱniacin,ȱvitaminȱB6,ȱvitaminȱB12,ȱandȱ
pantothenicȱacid.ȱȱ
4.6.1ȱThiamineȱ(VitaminȱB1)ȱ
Thiamineȱ isȱ widelyȱ foundȱ inȱ foods,ȱ includingȱ meat,ȱ
legumes,ȱ andȱ wholeȱ orȱ enrichedȱ grainȱ products.ȱ Theȱ
Recommendedȱ Dietaryȱ Allowanceȱ (RDA)8ȱ forȱ thiamineȱ
forȱ adultȱ malesȱ isȱ 1.2ȱ mg/dayȱ andȱ forȱ adultȱ females,ȱ 1.1ȱ
8ȱTheȱ Dietaryȱ Referenceȱ Intakesȱ (DRIs)ȱ areȱ nutrientȱ referenceȱ
valuesȱ thatȱ replaceȱ theȱ 1990ȱ Recommendedȱ Nutrientȱ Intakesȱ
(RNIs)ȱ inȱ Canadaȱ andȱ theȱ 1989ȱ Recommendedȱ Dietaryȱ
AllowancesȱinȱtheȱUnitedȱStatesȱ(HealthȱCanada,ȱhttp://www.hcȬ
sc.gc.ca/fnȬan/nutrition/reference/dri_usingȬutil_anrefȬeng.php).ȱ
TheȱDRIsȱincludeȱRecommendedȱDietaryȱAllowanceȱ(RDA)ȱandȱ
AdequateȱIntakeȱ(AI).ȱȱ
www.intechopen.com
!
mg/dayȱ (IOMȱ 1998).ȱ Basedȱ onȱ dataȱ fromȱ theȱ Canadianȱ
Communityȱ Healthȱ Surveyȱ (CCHS)ȱ (2004),ȱ theȱ 95thȱ
percentileȱofȱdietaryȱintakeȱindicatesȱthatȱadultsȱconsumeȱ
upȱ toȱ 4ȱ mgȱ ofȱ thiamineȱ perȱ day.ȱ Theȱ formulationȱ ofȱ aȱ
typicalȱenergyȱdrinkȱproductȱasȱdescribedȱinȱTableȱ1ȱdoesȱ
notȱ containȱ thiamine;ȱ however,ȱ otherȱ energyȱ drinksȱ mayȱ
containȱupȱtoȱ5ȱmgȱofȱthiamineȱperȱserving.ȱ
ȱ
TheȱUSȱInstituteȱofȱMedicineȱ(IOM)ȱwasȱnotȱableȱtoȱsetȱaȱ
tolerableȱupperȱintakeȱlevelȱ(UL)ȱforȱthiamineȱdueȱtoȱtheȱ
lackȱ ofȱ dataȱ onȱ adverseȱ effectsȱ (IOMȱ 1998).ȱ Inȱ addition,ȱ
theȱEuropeanȱ Commissionȱconcludedȱthatȱwhileȱitȱisȱnotȱ
possibleȱ toȱ deriveȱ aȱ ULȱ forȱ thiamine,ȱ currentȱ levelsȱ ofȱ
intakeȱfromȱallȱsourcesȱdoȱnotȱrepresentȱaȱhealthȱriskȱforȱ
theȱ generalȱ populationȱ (Europeanȱ Commissionȱ 2001).ȱ
However,ȱ theȱ Australiaȱ Newȱ Zealandȱ Foodȱ Authorityȱ
(ANZFA)ȱsetȱaȱmaximumȱlimitȱforȱthiamineȱofȱ20ȱmg/250ȱ
mlȱinȱformulatedȱcaffeinatedȱbeveragesȱasȱaȱconservativeȱ
limitȱthatȱwasȱbasedȱonȱaȱmaximumȱoneȬdayȱquantityȱofȱ
40ȱmgȱthiamineȱ(ANZFAȱ2001).ȱȱ
ȱ
Orallyȱ ingestedȱ thiamineȱ hasȱ aȱ longȱ historyȱ ofȱ useȱ asȱ aȱ
supplementȱ withoutȱ reportedȱ adverseȱ effects.ȱ Thereȱ areȱ
noȱ reportsȱ ofȱ adverseȱ effectsȱ ofȱ oralȱ thiamine,ȱ evenȱ atȱ
dosagesȱ ofȱ severalȱ hundredȱ milligramsȱ perȱ dayȱ
(Europeanȱ Commissionȱ 2001).ȱ Aȱ Canadianȱ evaluationȱ ofȱ
micronutrientȱsafetyȱclassifiedȱthiamineȱasȱaȱnutrientȱwithȱ
noȱknownȱadverseȱeffectsȱ(ProgramȱonȱFoodȱSafetyȱ1996).ȱȱ
4.6.2ȱRiboflavinȱ(VitaminȱB2)ȱ
Riboflavinȱ isȱ foundȱ inȱ aȱ wideȱ varietyȱ ofȱ foodsȱ butȱ milkȱ
andȱmilkȱproductsȱareȱthoughtȱtoȱcontributeȱtheȱmajorityȱ
ofȱ dietaryȱ riboflavin.ȱ Eggs,ȱ meat,ȱ andȱ legumesȱ alsoȱ
provideȱ riboflavinȱ inȱ significantȱ quantitiesȱ (Groffȱ &ȱ
Gropperȱ2000).ȱTheȱRDAȱforȱriboflavinȱforȱadultȱmalesȱisȱ
1.3ȱmg/dayȱandȱforȱadultȱfemales,ȱ1.1ȱmg/dayȱ(IOMȱ1998).ȱ
Basedȱ onȱ dataȱ fromȱ CCHSȱ (2004),ȱ theȱ 95thȱ percentileȱ ofȱ
dietaryȱintakeȱindicatesȱthatȱadultsȱconsumeȱupȱtoȱ4.5ȱmgȱ
ofȱ riboflavinȱ perȱ day.ȱ Theȱ typicalȱ energyȱ drinkȱ productȱ
formulation,ȱasȱdescribedȱinȱTableȱ1,ȱcontainsȱ1.65ȱmgȱofȱ
riboflavinȱ perȱ 250ȱ mlȱ servingȱ whileȱ otherȱ energyȱ drinksȱ
mayȱcontainȱupȱtoȱ5ȱmgȱofȱriboflavinȱperȱserving.ȱ
ȱ
IOMȱwasȱnotȱableȱtoȱsetȱaȱULȱforȱriboflavinȱdueȱtoȱtheȱlackȱofȱ
dataȱ onȱ adverseȱ effectsȱ (IOMȱ 1998).ȱ Inȱ addition,ȱ theȱ
Europeanȱ Commissionȱ concludedȱ thatȱ whileȱ itȱ isȱ notȱ
possibleȱtoȱderiveȱaȱULȱforȱriboflavin,ȱcurrentȱlevelsȱofȱintakeȱ
fromȱ allȱ sourcesȱ doȱ notȱ representȱ aȱ riskȱ toȱ humanȱ healthȱ
(Europeanȱ Commissionȱ 2000).ȱ However,ȱ ANZFAȱ setȱ aȱ
maximumȱ limitȱ forȱ riboflavinȱ ofȱ 20ȱ mg/dayȱ inȱ formulatedȱ
caffeinatedȱbeveragesȱbasedȱonȱtheȱcompositionȱofȱRedȱBullȱ
andȱknowledgeȱofȱregularȱconsumptionȱofȱ500ȱmlȱperȱdayȱofȱ
thisȱproductȱ(ANZFAȱ2001).ȱȱ
ȱ
Theȱ toxicityȱ ofȱ riboflavinȱ isȱ consideredȱ toȱ beȱ extremelyȱ
lowȱ dueȱ inȱ partȱ toȱ readyȱ excretionȱ ofȱ excessȱ amountsȱ
www.intechopen.com
ȱ
(ANZFAȱ 2001).ȱ Availableȱ subȬchronicȱ dataȱ fromȱ humanȱ
studiesȱ andȱ pharmacokineticȱ studiesȱ doȱ notȱ showȱ
reportedȱeffectsȱonȱoralȱtoxicityȱofȱriboflavin.ȱApartȱfromȱ
aȱ fewȱ minorȱ gastrointestinalȱ disorders,ȱ whichȱ areȱ notȱ
clearlyȱrelatedȱtoȱriboflavinȱintake,ȱitȱisȱfreeȱfromȱseriousȱ
sideȱeffectsȱ(EuropeanȱCommissionȱ2000).9,10ȱ
4.6.3ȱNiacinȱ(VitaminȱB3)ȱ
Niacinȱ isȱ theȱ termȱ usedȱ toȱ describeȱ vitaminȱ B3;ȱ nicotinicȱ
acidȱ andȱ nicotinamideȱ areȱ twoȱ differentȱ formsȱ ofȱ niacin.ȱ
Theȱ bestȱ dietaryȱ sourcesȱ ofȱ niacinȱ includeȱ tuna,ȱ beef,ȱ
otherȱ meats,ȱ andȱ cerealȱ grainsȱ (Groffȱ &ȱ Gropperȱ 2000).ȱ
TheȱRDAȱforȱniacinȱforȱadultȱmalesȱisȱ16ȱmg/dayȱandȱforȱ
adultȱfemales,ȱ14ȱmg/dayȱ(IOMȱ1998).ȱBasedȱonȱdataȱfromȱ
CCHSȱ (2004),ȱ theȱ 95thȱ percentileȱ ofȱ dietaryȱ intakeȱ
indicatesȱthatȱadultsȱconsumeȱupȱtoȱ76.7ȱmgȱofȱniacinȱperȱ
day.ȱ Theȱ typicalȱ energyȱ drink,ȱ asȱ definedȱ inȱ Tableȱ 1,ȱ
containsȱ 18ȱ mgȱ ofȱ niacinȱ perȱ 250ȱ mlȱ servingȱ whileȱ otherȱ
energyȱ drinksȱ mayȱ containȱ upȱ toȱ 40ȱ mgȱ ofȱ niacinȱ perȱ
serving.ȱTheȱniacinȱinȱtheseȱproductsȱisȱgenerallyȱpresentȱ
inȱtheȱformȱofȱnicotinamideȱandȱisȱmostȱlikelyȱaddedȱdueȱ
toȱitsȱroleȱinȱenergyȱmetabolism.ȱȱ
ȱ
Theȱ Instituteȱ ofȱ Medicineȱ (IOM)ȱ inȱ theȱ Unitedȱ Statesȱ
(1998)ȱ setȱ aȱ ULȱ ofȱ 35ȱ mg/dayȱ forȱ niacinȱ basedȱ onȱ theȱ
adverseȱ effectȱ ofȱ flushing11.ȱ Flushingȱ isȱ firstȱ observedȱ
afterȱ excessȱ niacinȱ intakeȱ andȱ isȱ generallyȱ observedȱ atȱ
lowerȱdosesȱthanȱareȱotherȱeffects.ȱFlushingȱthatȱresultsȱinȱ
patientsȱ decidingȱ toȱ changeȱ theȱ patternȱ ofȱ niacinȱ intakeȱ
(i.e.,ȱreduceȱtheȱamountȱtakenȱatȱaȱtimeȱorȱwithdrawȱfromȱ
treatment)ȱwasȱselectedȱasȱtheȱmostȱappropriateȱendpointȱ
onȱ whichȱ toȱ baseȱ aȱ UL.ȱ Althoughȱ nicotinamideȱ appearsȱ
notȱ toȱ beȱ associatedȱ withȱ flushingȱ effects,ȱ aȱ ULȱ forȱ
nicotinicȱ acidȱ thatȱ isȱ basedȱ onȱ flushingȱ isȱ consideredȱ
protectiveȱ againstȱ potentialȱ adverseȱ effectsȱ ofȱ
nicotinamideȱ(IOMȱ1998).ȱ
ȱ
Otherȱ jurisdictionsȱ haveȱ setȱ ULsȱ forȱ bothȱ nicotinicȱ acidȱ
andȱ nicotinamide.ȱ However,ȱ asȱ nicotinamideȱ isȱ typicallyȱ
foundȱinȱenergyȱdrinks,ȱonlyȱtheȱULsȱrelatingȱtoȱthisȱformȱ
ofȱ niacinȱ areȱ discussed.ȱ Theȱ Europeanȱ Commission set a UL ofȱ900ȱmg/dayȱ(12.5ȱmg/kgȱbw/day)ȱforȱ nicotinamideȱ
basedȱ onȱ aȱ NOAELȱ ofȱ 25ȱ mg/kgȱ bw/d,ȱ derivedȱ fromȱ
studiesȱ inȱ subjectsȱ withȱ orȱ atȱ riskȱ ofȱ diabetes.ȱ (Europeanȱ
Commission,ȱ2002).ȱȱ
ȱ
9ȱ Whileȱ thisȱ manuscriptȱ wasȱ inȱ preparation,ȱ Healthȱ Canadaȱ
publishedȱ theȱ documentȱ Categoryȱ Specificȱ Guidanceȱ forȱ
TemporaryȱMarketingȱAuthorization:ȱCaffeinatedȱEnergyȱDrinksȱ
(Marchȱ 2012).ȱ Aȱ dailyȱ maximumȱ levelȱ ofȱ 5ȱ mg/dayȱ wasȱ setȱ forȱ
theȱadditionȱofȱthiamineȱtoȱenergyȱdrinks.ȱ
11ȱ Healthȱ Canada’sȱ Categoryȱ Specificȱ Guidanceȱ forȱ Temporaryȱ
Marketingȱ Authorization:ȱ Caffeinatedȱ Energyȱ Drinksȱ (Marchȱ
2012)ȱsetȱaȱdailyȱmaximumȱlevelȱofȱ27ȱmg/dayȱforȱtheȱadditionȱofȱ
riboflavinȱtoȱenergyȱdrinks.ȱ
ȱ
11
ȱFlushingȱrefersȱtoȱaȱtransitoryȱsensationȱofȱextremeȱheat.ȱ
15
!
Theȱ Expertȱ Groupȱ onȱ Vitaminsȱ andȱ Mineralsȱ (Foodȱ
Standardsȱ Agencyȱ 2003)ȱ setȱ aȱ guidanceȱ levelȱ ofȱ 560ȱ
mg/dayȱ forȱ nicotinamideȱ asȱ theȱ limitedȱ dataȱ onȱ theȱ
occurrenceȱ ofȱ nicotinamideȱ toxicityȱ indicatesȱ thatȱ itȱ isȱ
quiteȱlow.ȱȱ
ȱ
ANZFAȱ setȱ aȱ maximumȱ limitȱ forȱ niacinȱ ofȱ 40ȱ mg/dayȱ (2ȱ
cansȱ ofȱ 20ȱ mg/250ȱ ml)ȱ inȱ formulatedȱ caffeinatedȱ
beveragesȱbasedȱonȱtheirȱassessmentȱ(ANZFAȱ2001).12ȱ
ȱ
Thereȱ isȱ noȱ evidenceȱ ofȱ adverseȱ effectsȱ fromȱ theȱ
consumptionȱofȱnormalȱlevelsȱofȱniacinȱinȱfoods.ȱAdverseȱ
effectsȱ haveȱ beenȱ observedȱ withȱ intakesȱ ofȱ nicotinamideȱ
greaterȱ thanȱ 3000ȱ mg/dayȱ comparedȱ withȱ intakesȱ ofȱ
nicotinicȱ acidȱ ofȱ 1500ȱ mg/dayȱ (ANZFAȱ 2001).ȱ Adverseȱ
effectsȱcanȱbeȱobservedȱfollowingȱhighȱintakesȱofȱnicotinicȱ
acid,ȱ whichȱ mayȱ beȱ achievedȱ throughȱ consumptionȱ ofȱ
pharmacologicalȱ preparationsȱ orȱ dietaryȱ supplementalȱ
products.ȱ Adverseȱ effectsȱ associatedȱ withȱ theȱ useȱ ofȱ
nicotinicȱacidȱasȱaȱdrug,ȱespeciallyȱinȱdosesȱofȱ1ȱgȱorȱmoreȱ
perȱdayȱinclude:ȱ
ȱ
x
Possibleȱ injuryȱ toȱ theȱ liver,ȱ asȱ indicatedȱ byȱ elevatedȱ
serumȱ levelsȱ ofȱ enzymesȱ ofȱ hepaticȱ originȱȱ
(e.g.,ȱ transaminases)ȱ andȱ byȱ obstructionȱ ofȱ normalȱ
bileȱflowȱfromȱtheȱliverȱtoȱtheȱsmallȱintestine;ȱ
x
Developmentȱ ofȱ dermatologicalȱ problemsȱ suchȱ asȱ
itching;ȱandȱ
x
Elevationȱ ofȱ plasmaȱ glucoseȱ levelsȱ (Groffȱ andȱ
Gropperȱ2000).ȱ
4.6.4ȱPyridoxineȱ(VitaminȱB6)ȱȱ
Excellentȱ sourcesȱ ofȱ vitaminȱ B6ȱ inȱ commonlyȱ consumedȱ
foodsȱ areȱ bananas,ȱ navyȱ beans,ȱ andȱ walnutsȱ (Groffȱ &ȱ
Gropperȱ2000).ȱTheȱRDAȱforȱvitaminȱB6ȱforȱadultȱmalesȱisȱ
1.3Ȭ1.7ȱ mg/dayȱ andȱ forȱ adultȱ females,ȱ 1.3Ȭ1.5ȱ mg/dayȱ
(IOMȱ 1998).ȱ Basedȱ onȱ dataȱ fromȱ CCHSȱ (2004),ȱ theȱ 95thȱ
percentileȱofȱdietaryȱintakeȱindicatesȱthatȱadultsȱconsumeȱ
upȱ toȱ 4.5ȱ mgȱ ofȱ vitaminȱ B6ȱ perȱ day.ȱ Theȱ typicalȱ energyȱ
drink,ȱ asȱ describedȱ inȱ Tableȱ 1,ȱ containsȱ 2ȱ mgȱ ofȱ vitaminȱ
B6ȱ perȱ 250ȱ mlȱ servingȱ whileȱ otherȱ energyȱ drinksȱ mayȱ
containȱupȱtoȱ10ȱmgȱofȱvitaminȱB6ȱperȱserving.ȱ
ȱ
IOMȱ setȱ aȱ ULȱ forȱ vitaminȱ B6ȱ ofȱ 100ȱ mg/dayȱ basedȱ onȱ aȱ
NOAELȱ ofȱ 200ȱ mg/dayȱ andȱ applyingȱ anȱ uncertaintyȱ
factorȱ ofȱ 2ȱ toȱ accountȱ forȱ theȱ limitationsȱ inȱ dataȱ (IOMȱ
1998).ȱTheȱEuropeanȱCommissionȱsetȱaȱULȱofȱ25ȱmg/dayȱ
forȱ vitaminȱ B6ȱ (Europeanȱ Commissionȱ 2000)ȱ byȱ takingȱ
theȱaverageȱintakesȱofȱvitaminȱB6ȱ(100ȱmg)ȱfromȱtheȱstudyȱ
byȱDaltonȱandȱDaltonȱ(1987)ȱandȱapplyingȱanȱuncertaintyȱ
factorȱofȱ4ȱtoȱaccountȱforȱlongȬtermȱintake.ȱȱ
ȱ Healthȱ Canada’sȱ Categoryȱ Specificȱ Guidanceȱ forȱ Temporaryȱ
2012)ȱsetȱaȱdailyȱmaximumȱlevelȱofȱ450ȱmg/dayȱforȱtheȱadditionȱ
ofȱnicotinamideȱtoȱenergyȱdrinks.ȱ
12
ȱ
ȱ
ANZFAȱsetȱaȱmaximumȱlimitȱforȱvitaminȱB6ȱofȱ10ȱmg/dayȱ
inȱ formulatedȱ caffeinatedȱ beveragesȱ basedȱ onȱ historyȱ ofȱ
use,ȱratherȱthanȱtheȱU.S.ȱULȱ(ANZFAȱ2001).ȱȱ
ȱ
Thereȱ areȱ noȱ safetyȱ concernsȱ inȱ relationȱ toȱ vitaminȱ B6ȱ
intakeȱfromȱfoodȱsources.ȱHowever,ȱadverseȱneurologicalȱ
effectsȱ haveȱ beenȱ detectedȱ inȱ humansȱ afterȱ veryȱ highȱ
dosesȱ (>500ȱ mg/day,ȱ equivalentȱ toȱ approximatelyȱ 8ȱ
mg/kg/day)13.ȱ Minorȱ neurologicalȱ symptomsȱ mayȱ beȱ
apparentȱatȱdosesȱofȱ100ȱmg/dayȱorȱmoreȱifȱconsumedȱforȱ
longȱ periods.ȱ Thereȱ areȱ noȱ subgroupsȱ thatȱ areȱ knownȱ toȱ
beȱunusuallyȱsusceptibleȱtoȱtheȱadverseȱeffectsȱofȱvitaminȱ
B6ȱ(EuropeanȱCommissionȱ2000).ȱ
4.6.5ȱCobalaminȱ(VitaminȱB12)ȱȱ
Theȱ onlyȱ dietaryȱ sourcesȱ ofȱ vitaminȱ B12ȱ forȱ humansȱ areȱ
fromȱ animalȱ products,ȱ whichȱ haveȱ derivedȱ theirȱ
cobalaminsȱ fromȱ microorganisms.ȱ Theȱ bestȱ sourcesȱ ofȱ
cobalaminsȱ areȱ meatȱ andȱ meatȱ products,ȱ poultryȱ andȱ
eggs.ȱ Theȱ RDAȱ forȱ vitaminȱ B12ȱ forȱ adultȱ malesȱ andȱ
femalesȱ isȱ 2.4ȱ mcg/dayȱ (micrograms/day)ȱ (IOMȱ 1998).ȱ
Basedȱ onȱ dataȱ fromȱ CCHSȱ (2004),ȱ theȱ 95thȱ percentileȱ ofȱ
dietaryȱintakeȱindicatesȱthatȱadultsȱconsumeȱupȱtoȱ6ȱmcgȱ
ofȱ vitaminȱ B12ȱ perȱ day.ȱ Theȱ typicalȱ amountȱ usedȱ forȱ
energyȱdrinks,ȱasȱdescribedȱinȱTableȱ1,ȱcontainsȱ1ȱmcgȱofȱ
vitaminȱB12ȱperȱ250ȱmlȱservingȱwhileȱotherȱenergyȱdrinksȱ
mayȱcontainȱupȱtoȱ20ȱmcgȱofȱvitaminȱB12ȱperȱserving.ȱ
ȱ
IOMȱ wasȱ notȱ ableȱ toȱ setȱ aȱ ULȱ forȱ vitaminȱ B12ȱ dueȱ toȱ theȱ
lackȱ ofȱ dataȱ onȱ adverseȱ effectsȱ (IOMȱ 1998).ȱ Similarly,ȱ theȱ
EuropeanȱCommissionȱconcludedȱthatȱitȱisȱnotȱpossibleȱtoȱ
deriveȱaȱULȱforȱvitaminȱB12ȱasȱthereȱareȱnoȱclearlyȱdefinedȱ
adverseȱ effectsȱ producedȱ fromȱ thisȱ vitaminȱ (Europeanȱ
Commissionȱ 2000).ȱ ANZFAȱ setȱ aȱ maximumȱ limitȱ forȱ
vitaminȱ B12ȱ ofȱ 10ȱ mcg/dayȱ inȱ formulatedȱ caffeinatedȱ
beveragesȱbasedȱonȱhistoryȱofȱuseȱ(ANZFAȱ2001).ȱȱ
ȱ
Noȱ adverseȱ effectsȱ haveȱ beenȱ associatedȱ withȱ excessȱ
vitaminȱ B12ȱ intakeȱ fromȱ foodȱ orȱ supplementsȱ inȱ healthyȱ
individualsȱ(EuropeanȱCommissionȱ2000).ȱ14ȱ
4.6.6ȱPantothenicȱacidȱ(VitaminȱB5)ȱȱ
Meatsȱ (especiallyȱ liver),ȱ eggȱ yolk,ȱ legumes,ȱ andȱ wholeȱ
grainȱ cerealsȱ areȱ goodȱ sourcesȱ ofȱ pantothenicȱ acid.ȱ Theȱ
adequateȱintakeȱ(AI)ȱforȱpantothenicȱacidȱforȱadultȱmalesȱ
andȱfemalesȱisȱ5ȱmg/dayȱ(IOMȱ1998).ȱCCHSȱ(2004)ȱdidȱnotȱ
haveȱ anyȱ dietaryȱ intakeȱ dataȱ onȱ pantothenicȱ acid;ȱ
however,ȱanotherȱsourceȱindicatesȱthatȱaverageȱintakesȱofȱ
adultsȱ rangeȱ betweenȱ 3Ȭ12ȱ mg/dȱ (Europeanȱ Commissionȱ
2002).ȱ Theȱ typicalȱ energyȱ drink,ȱ asȱ definedȱ inȱ Tableȱ 1,ȱ
ȱAtȱdosesȱlessȱthanȱ500ȱmg/day,ȱtheȱadverseȱeffectsȱwereȱreversible.ȱ
2012)ȱsetȱaȱdailyȱmaximumȱlevelȱofȱ25ȱmcg/dayȱforȱtheȱadditionȱ
ofȱvitaminȱB12ȱtoȱenergyȱdrinks.ȱ
13
14
www.intechopen.com
!
containsȱ 6ȱ mgȱ ofȱ pantothenicȱ acidȱ perȱ 250ȱ mlȱ servingȱ
whileȱ otherȱ energyȱ drinksȱ mayȱ containȱ upȱ toȱ 10ȱ mgȱ ofȱ
pantothenicȱacidȱperȱserving.ȱ
ȱ
IOMȱwasȱnotȱableȱtoȱsetȱaȱULȱforȱpantothenicȱacidȱdueȱtoȱ
theȱlackȱofȱdataȱonȱadverseȱeffectsȱ(IOMȱ1998).ȱInȱaddition,ȱ
theȱ Europeanȱ Commissionȱ concludedȱ thatȱ whileȱ itȱ isȱ notȱ
possibleȱtoȱderiveȱaȱULȱforȱpantothenicȱacid,ȱcurrentȱlevelsȱ
ofȱintakeȱfromȱallȱsourcesȱdoȱnotȱrepresentȱaȱhealthȱriskȱforȱ
theȱ generalȱ populationȱ (Europeanȱ Commissionȱ 2000).ȱ
ANZFAȱsetȱaȱmaximumȱlimitȱofȱ10ȱmg/dayȱforȱpantothenicȱ
acidȱ inȱ formulatedȱ caffeinatedȱ beveragesȱ basedȱ onȱ theȱ
compositionȱofȱtheȱreferenceȱproductȱknownȱasȱRedȱBull™ȱ
andȱknowledgeȱofȱregularȱconsumptionȱofȱ500ȱmLȱperȱdayȱ
ofȱthisȱproductȱ(ANZFAȱ2001).ȱ
ȱ
Thereȱ areȱ noȱ reportsȱ ofȱ pantothenicȱ acidȱ orȱ panthenolȱ
toxicityȱinȱhumans.ȱMinorȱgastrointestinalȱeffectsȱsuchȱasȱ
occasionalȱdiarrheaȱandȱwaterȱretentionȱoccurredȱonlyȱatȱ
veryȱ highȱ intakesȱ (10Ȭ20ȱ g/day)ȱ (Europeanȱ Commissionȱ
2002).15ȱ
4.7ȱSummaryȱȱ
Healthȱhazardȱdataȱonȱenergyȱdrinksȱareȱextremelyȱlimitedȱ
andȱ thereforeȱ theȱ hazardȱ assessmentȱ wasȱ basedȱ onȱ
individualȱ ingredients.ȱ Caffeineȱ wasȱ identifiedȱ asȱ theȱ
ingredientȱ inȱ energyȱ drinksȱ havingȱ theȱ greatestȱ potentialȱ
forȱ intakesȱ ofȱ healthȱ concern.ȱ Assumingȱ noȱ additionalȱ
caffeineȱfromȱtheȱdiet,ȱitȱwasȱdeterminedȱthatȱnoȱmoreȱthanȱ
5ȱ servingsȱ perȱ dayȱ ofȱ aȱ typicalȱ energyȱ drinkȱ shouldȱ beȱ
consumedȱbyȱtheȱgeneralȱadultȱpopulation.ȱAtȱthisȱlevelȱofȱ
consumption,ȱtheȱlevelsȱofȱtaurineȱandȱglucuronolactoneȱinȱ
aȱ typicalȱ energyȱ drinkȱ areȱ notȱ expectedȱ toȱ poseȱ aȱ healthȱ
hazardȱ inȱ theȱ shortȱ term.ȱ Althoughȱ thereȱ areȱ limitedȱ
hazardȱ dataȱ onȱ energyȱ drinksȱ asȱ formulatedȱ inȱ theȱ
publishedȱliterature,ȱactualȱuseȱofȱenergyȱdrinksȱhasȱbeenȱ
associatedȱwithȱsomeȱadverseȱreactions.ȱHowever,ȱdueȱtoȱ
theȱ absenceȱ ofȱ longȬtermȱ safetyȱ dataȱ onȱ highȱ levelsȱ ofȱ
consumptionȱ ofȱ taurineȱ andȱ glucuronolactone,ȱ theȱ
potentialȱ interactionȱ ofȱ theseȱ substancesȱ withȱ caffeine,ȱ
andȱ forȱ mostȱ consumers,ȱ theȱ knownȱ additionȱ ofȱ caffeineȱ
fromȱ otherȱ dietaryȱ sources,ȱ itȱ wasȱ concludedȱ thatȱ theȱ
longȬtermȱconsumptionȱofȱ5ȱservingsȱofȱenergyȱdrinksȱperȱ
dayȱ couldȱ notȱ beȱ consideredȱ toȱ representȱ noȱ healthȱ
concernȱforȱtheȱgeneralȱadultȱpopulation.ȱȱ
ȱ
Also,ȱtheȱevidenceȱexaminedȱ suggestsȱthatȱ mostȱofȱtheȱBȱ
vitaminsȱandȱotherȱconstituentsȱofȱaȱtypicalȱenergyȱdrinkȱ
wouldȱ notȱ poseȱ aȱ healthȱ hazardȱ inȱ theȱ shortȱ term,ȱ butȱ
longȬtermȱ safetyȱ dataȱ wereȱ notȱ availableȱ forȱ thisȱ levelȱofȱ
consumption.ȱ
www.intechopen.com
ȱ
5.ȱSurveysȱandȱStudiesȱonȱEnergyȱDrinkȱȱ
ConsumptionȱLevelsȱandȱPatternsȱ
5.1ȱIntroductionȱ
Aȱlimitedȱnumberȱofȱqualitativeȱandȱquantitativeȱsurveysȱ
andȱ studiesȱ onȱ energyȱ drinkȱ consumptionȱ haveȱ beenȱ
carriedȱ outȱ inȱ otherȱ jurisdictionsȱ sinceȱ theȱ lateȱ 1990’s.ȱ
Dataȱ onȱ levelsȱ ofȱ intakeȱ asȱ wellȱ asȱ patternsȱ ofȱ
consumptionȱ wereȱ collectedȱ forȱ variousȱ ageȱ groupsȱ asȱ
describedȱinȱtheȱsurveyȱandȱstudyȱsummariesȱbelow.ȱȱ
ȱInȱmales,ȱthisȱwouldȱresultȱinȱaȱdailyȱdoseȱofȱ5.5ȱmg/kgȱbwȱ(=ȱ
441ȱmgȱofȱcaffeine/ȱ80ȱkgȱbwȱ=ȱ281ȱmgȱofȱcaffeineȱ+ȱ160ȱmgȱfromȱ
2ȱservingsȱofȱenergyȱdrinksȱ/ȱ80ȱkgȱbw)ȱandȱinȱfemalesȱinȱaȱdailyȱ
doseȱofȱ6.0ȱmg/kgȱbwȱ(=ȱ390ȱmgȱofȱcaffeineȱ/ȱ65ȱkgȱbwȱ=ȱ230ȱmgȱ
ofȱcaffeineȱ+ȱ160ȱmgȱfromȱ2ȱservingsȱofȱenergyȱdrinksȱ/ȱ65ȱkgȱbw).ȱ
16
2012)ȱsetȱaȱdailyȱmaximumȱlevelȱofȱ50ȱmg/dayȱforȱtheȱadditionȱofȱ
pantothenicȱacidȱtoȱenergyȱdrinks.ȱȱ
15
Theȱhealthȱhazardȱassessmentȱconcludedȱthatȱtheȱgeneralȱ
adultȱ populationȱ couldȱ consumeȱ 2ȱ servingsȱ ofȱ aȱ typicalȱ
energyȱ drinkȱ perȱ dayȱ withȱ noȱ expectedȱ negativeȱ healthȱ
consequences.ȱThisȱconclusionȱwasȱbasedȱonȱtheȱsafetyȱofȱ
theȱnonȬcaffeineȱingredientsȱofȱenergyȱdrinksȱ(i.e.ȱtaurine,ȱ
glucuronolactone,ȱinositolȱandȱBȱvitamins)ȱatȱthisȱlevelȱofȱ
consumption,ȱandȱtheȱfactȱthatȱcaffeineȱfromȱotherȱdietaryȱ
sourcesȱinȱadditionȱtoȱthatȱinȱ2ȱservingsȱofȱenergyȱdrinksȱ
wouldȱ notȱ poseȱ aȱ healthȱ riskȱ toȱ theȱ generalȱ adultȱ
population.ȱȱ
ȱ
Moreȱ specifically,ȱ theȱ respectiveȱ meanȱ dailyȱ intakesȱ ofȱ
caffeineȱ fromȱ allȱ dietaryȱ sourcesȱ forȱ adultȱ Canadianȱ
malesȱ andȱ femalesȱ areȱ 281ȱ andȱ 230ȱ mg.ȱ Withȱ bodyȱ
weightsȱofȱ80ȱandȱ65ȱkg,ȱthesesȱintakesȱareȱequivalentȱtoȱ
aȱ dailyȱ doseȱ ofȱ 3.5ȱ andȱ 3.1ȱ mg/kgȱ bw,ȱ respectivelyȱ
(Statisticsȱ Canada,ȱ Canadianȱ Communityȱ Healthȱ
Survey,ȱ 2004).ȱ Theȱ consumptionȱ ofȱ twoȱ servingsȱ ofȱ aȱ
typicalȱ energyȱ drinkȱ containingȱ 80ȱ mgȱ ofȱ caffeineȱ perȱ
servingȱ wouldȱ resultȱinȱ theȱadditionȱ ofȱ 160ȱ mgȱ caffeineȱ
toȱtheȱdiet.ȱInȱmales,ȱthisȱwouldȱresultȱinȱaȱdailyȱdoseȱofȱ
5.5ȱmg/kgȱbwȱandȱinȱfemales,ȱaȱdailyȱdoseȱofȱ6.0ȱmg/kgȱ
bw.16ȱ Healthȱ Canada’sȱ recommendedȱ maximumȱ dailyȱ
intakeȱofȱcaffeineȱforȱadultsȱisȱ6.5ȱmg/kgȱbwȱorȱaboutȱ400ȱ
mgȱ forȱ anȱ adultȱ weighingȱ 65ȱ kgȱ (Nawrotȱ etȱ al.,ȱ 2003).ȱ
Givenȱ thatȱ theȱ additionȱ ofȱ twoȱ servingsȱ ofȱ aȱ typicalȱ
energyȱ drinkȱ toȱ theȱ dietȱ wouldȱ notȱ exceedȱ theȱ
recommendedȱ maximumȱ dailyȱ intakeȱ ofȱ caffeine,ȱ itȱ canȱ
beȱ concludedȱ thatȱ thisȱ levelȱ ofȱ consumptionȱ wouldȱ notȱ
poseȱ anȱ additionalȱ healthȱ hazardȱ basedȱ onȱ theȱ caffeineȱ
contentȱofȱtheseȱproducts.ȱȱ
ȱ
Theȱ consumptionȱ ofȱ energyȱ drinksȱ byȱ subpopulations,ȱ
suchȱasȱchildrenȱandȱpregnantȱandȱbreastfeedingȱwomen,ȱ
isȱnotȱgenerallyȱrecommended.ȱBasedȱonȱcaffeineȱcontent,ȱ
consumptionȱ ofȱ suchȱ drinksȱ byȱ anyȱ groupȱ shouldȱ beȱ
limitedȱ toȱ theirȱ recommendedȱ maximumȱ dailyȱ intakeȱ ofȱ
caffeine.ȱ Excessȱ consumptionȱ ofȱ energyȱ drinksȱ wouldȱ beȱ
expectedȱtoȱresultȱinȱhealthȱconsequencesȱsimilarȱtoȱthoseȱ
fromȱexcessȱexposureȱtoȱcaffeine.ȱȱ
17
!
5.2ȱInternationalȱsurveyȱdataȱ
DataȱfromȱtheȱEuropeanȱCommissionȱ(ECȱ1999)ȱindicatedȱ
thatȱ 9%ȱ ofȱ theȱ populationȱ couldȱ beȱ describedȱ asȱ ‘regularȱ
consumers’ȱ ofȱ ȱ energyȱ drinks,ȱ withȱ intakeȱ byȱ theseȱ
consumersȱlikelyȱtoȱbeȱinȱtheȱorderȱofȱ500ȱmlȱperȱdayȱ(2ȱxȱ
250ȱmlȱcans).17ȱ
ȱ
Anȱ Australianȱ surveyȱ assessedȱ theȱ incidenceȱ ofȱ
consumptionȱofȱenergyȱdrinksȱwithinȱaȱtwoȬweekȱperiodȱ
inȱ 1999.ȱ However,ȱ thisȱ studyȱ didȱ notȱ recordȱ quantitiesȱ
consumedȱorȱfrequencyȱofȱconsumption.ȱTheȱresultsȱfromȱ
thisȱsurveyȱareȱgivenȱinȱTableȱ7.ȱ
ȱ
Sampleȱ
141ȱȱ
240ȱ
%ȱConsumersȱ
27%ȱmalesȱagedȱ8Ȭ12ȱyearsȱ
12%ȱfemalesȱagedȱ8Ȭ12ȱyearsȱ
24%ȱmalesȱagedȱ12Ȭȱ18ȱyearsȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱȱ20%ȱfemalesȱ
agedȱ12Ȭȱ18ȱyearsȱ
Tableȱ7.ȱEnergyȱdrinkȱconsumptionȱinȱAustraliaȱinȱaȱTwoȱWeekȱ
Periodȱduringȱ1999ȱ
Newȱ Zealandȱ dataȱ areȱ availableȱ fromȱ theȱ Panoramaȱ
surveyȱ ofȱ 12ȱ 000ȱ respondentsȱ agedȱ 10ȱ yearsȱ andȱ over,ȱ
conductedȱ byȱ ACNielsenȱ inȱ 1999.ȱ Twoȱ hundredȱ andȱ
sixtyȬfourȱ (264)ȱ individualsȱ agedȱ 10ȱ toȱ 14ȱ yearsȱ wereȱ
interviewedȱ regardingȱ theirȱ consumptionȱ ofȱ energyȱ
drinks,ȱ andȱ theȱ resultantȱ dataȱ forȱ theȱ 64ȱ (24%)ȱ whoȱ
reportedȱdrinkingȱthemȱareȱpresentedȱinȱTableȱ8.ȱȱ
ȱ
Consumptionȱrateȱ
No.ȱ
consumersȱȱ
Atȱleastȱonceȱperȱdayȱ
6ȱ
Onceȱaȱweekȱȱ
21ȱ
Atȱleastȱonceȱaȱ
37ȱ
monthȱȱ
Percentageȱ10Ȭ14ȱyearȱ
oldsȱ
2ȱ
8ȱ
4ȱ
Tableȱ8.ȱEnergyȱdrinkȱconsumptionȱinȱNewȱZealandȱbyȱ10ȱtoȱ14ȱ
yearȱoldsȱ
Inȱ 2001,ȱ Redȱ Bullȱ Gmbhȱ conductedȱ aȱ surveyȱ ofȱ energyȱ
drinkȱconsumptionȱinȱAustria.ȱTheȱsurveyȱwasȱconductedȱ
onȱ 8500ȱ Austriansȱ agedȱ 15ȱ yearsȱ andȱ over.ȱ FortyȬtwoȱ
percentȱ ofȱ theȱ sampleȱ consumedȱ energyȱ drinksȱ atȱ leastȱ
occasionallyȱ andȱ 12%ȱ wereȱ regularȱ usersȱ whichȱ wereȱ
definedȱasȱthoseȱwhoȱconsumedȱatȱleastȱoneȱenergyȱdrinkȱ
perȱ week.ȱ Forȱ theȱ surveyedȱ population,ȱ meanȱ chronicȱ
consumption,ȱchronicȱconsumptionȱatȱtheȱ 95thȱpercentile,ȱ
andȱ acuteȱ consumptionȱ atȱ theȱ 90thȱ percentileȱ wereȱ
approximatelyȱ0.47ȱcansȱperȱday,ȱ1.4ȱcansȱperȱdayȱandȱ2.6ȱ
cansȱperȱday,ȱrespectivelyȱ(1ȱcanȱ=ȱ250ȱmL).ȱ
ȱ
ȱWhileȱthisȱmanuscriptȱwasȱinȱpreparation,ȱtheȱEuropeanȱFoodȱ
Safetyȱ Authorityȱ (EFSA)ȱ releasedȱ aȱ study,ȱ Gatheringȱ
consumptionȱdataȱonȱspecificȱconsumerȱgroupsȱofȱenergyȱdrinks.ȱ
(Zucconiȱetȱal.,ȱ2013),ȱdetailingȱtheȱconsumptionȱofȱenergyȱdrinksȱ
byȱEuropeans.ȱȱ
17
ȱ
Aȱmarketȱresearchȱsurveyȱofȱ1260ȱpeopleȱagedȱ11Ȭ35ȱyearsȱ
wasȱ conductedȱ inȱ Northernȱ Irelandȱ andȱ theȱ Republicȱ ofȱ
Irelandȱinȱ2001.ȱTheȱresultsȱofȱtheȱsurveyȱwereȱsimilarȱtoȱ
thoseȱ obtainedȱ inȱ theȱ Austrianȱ survey.ȱ Inȱ Northernȱ
IrelandȱandȱRepublicȱofȱIrelandȱrespectively,ȱ51%ȱpercentȱ
andȱ 37%ȱ ofȱ participantsȱ reportedȱ ‘ever’ȱ consumingȱ
energyȱ drinksȱ andȱ 10%ȱ andȱ 11%ȱ reportedȱ consumingȱ
energyȱ drinksȱ frequently.ȱ Amongȱ ‘ever’ȱ consumers,ȱ
averageȱ consumptionȱ wasȱ 3ȱ (250mL)ȱ cans/weekȱ andȱ forȱ
theȱ 95thȱ percentileȱ consumers,ȱ itȱ wasȱ 8ȱ cans/week.ȱ Theȱ
mostȱnumberȱofȱcansȱconsumedȱinȱaȱsingleȱsessionȱamongȱ
‘ever’ȱ consumersȱ averagedȱ approximatelyȱ 3ȱ cans,ȱ risingȱ
toȱ 8ȱ cansȱ amongȱ theȱ highestȱ consumers;ȱ theȱ comparableȱ
figureȱforȱ11Ȭ14ȱyearȬoldsȱwasȱapproximatelyȱ2ȱcans.ȱȱ
ȱ
Inȱ theȱ Irelandȱ survey,ȱ energyȱ drinkȱ consumptionȱ wasȱ
stronglyȱ relatedȱ toȱ alcoholȱ consumption.ȱ Amongȱ thoseȱ
whoȱ everȱ consumedȱ energyȱ drinks,ȱ 84%ȱ ofȱ allȱ
respondentsȱhadȱconsumedȱtheȱdrinkȱwithȱalcohol,ȱwhileȱ
56%ȱ reportedȱ consumingȱ theȱ drinksȱ withȱ alcoholȱ
regularly.ȱȱȱ
ȱ
O’Deaȱ(2003)ȱconductedȱaȱstudyȱtoȱobtainȱqualitativeȱdataȱ
aboutȱ theȱ typeȱ ofȱ nutritionalȱ supplementsȱ andȱ drinksȱ
consumedȱ byȱ adolescentsȱ inȱ Australia.ȱ SemiȬstructuredȱ
focusȱ groupȱ interviewsȱ wereȱ conductedȱ amongȱ 78ȱ
participantsȱ agedȱ 11Ȭ18ȱ years.ȱ Participantsȱ reportedȱ
consumingȱ sportȱ drinks,ȱ aȱ numberȱ ofȱ nutritionalȱ
supplementsȱ andȱ energyȱ drinks.ȱ Resultsȱ ofȱ theȱ studyȱ
indicatedȱ thatȱ adolescentsȱ appearȱ toȱ beȱ incorrectlyȱ
attributingȱ theȱ ‘creation’ȱ ofȱ energyȱ toȱ theseȱ supplementsȱ
whenȱ theȱ actualȱ effectȱ isȱ aȱ stimulantȱ effectȱ causedȱ byȱ
caffeine,ȱandȱotherȱstimulantsȱinȱproductsȱsuchȱasȱenergyȱ
drinks.ȱ Theȱ authorȱ notedȱ thatȱ manyȱ ofȱ theȱ misguidedȱ
‘energyȱcreation’ȱbeliefsȱmayȱbeȱattributedȱtoȱinformationȱ
providedȱ toȱ consumersȱ inȱ advertisingȱ andȱ marketingȱ ofȱ
theseȱ products.ȱ Accordingȱ toȱ theȱ author,ȱ theȱ adolescentsȱ
inȱ theȱ studyȱ hadȱ deliberatelyȱ soughtȱ anȱ ‘energyȱ boost’,ȱ
andȱhadȱreceivedȱitȱinȱtheȱformȱofȱaȱstimulantȱeffectȱfromȱ
theȱcaffeineȬcontainingȱsupplementsȱandȱdrinks.ȱȱ
ȱ
Malinauskasȱ etȱ al.ȱ (2007)ȱ conductedȱ aȱ surveyȱ ofȱ energyȱ
drinkȱconsumptionȱpatternsȱamongȱUnitedȱStatesȱcollegeȱ
studentsȱwhereinȱconsumptionȱpatternsȱofȱ496ȱrandomlyȱ
surveyedȱ collegeȱ studentsȱ wereȱ assessed.ȱ Fiftyȱ oneȱ
percentȱ ofȱ thoseȱ surveyedȱ reportedȱ consumingȱ greaterȱ
thanȱ oneȱ energyȱ drinkȱ eachȱ monthȱ inȱ anȱ averageȱ monthȱ
forȱtheȱsemester.ȱTheȱmajorityȱofȱusersȱconsumedȱenergyȱ
drinksȱ forȱ insufficientȱ sleepȱ (67%),ȱ toȱ increaseȱ energyȱ
(65%),ȱ andȱ toȱ drinkȱ withȱ alcoholȱ whileȱ partyingȱ (54%).ȱ
Theȱmajorityȱofȱusersȱconsumedȱoneȱenergyȱdrinkȱinȱmostȱ
situations,ȱ althoughȱ consumingȱ threeȱ orȱ moreȱ withȱ
alcoholȱwhileȱpartyingȱwasȱaȱcommonȱpracticeȱ(49%).ȱForȱ
theȱ majorityȱ ofȱ studentsȱ (73Ȭ86%),ȱ energyȱ drinksȱ wereȱ
consumedȱ 1Ȭ4ȱ daysȱ inȱ aȱ month.ȱ Fiveȱ toȱ 12%ȱ ofȱ studentsȱ
consumedȱenergyȱdrinksȱelevenȱorȱmoreȱdaysȱaȱmonth.ȱ
www.intechopen.com
!
6.ȱModellingȱExposureȱScenarioȱȱ
Theȱhealthȱriskȱassessmentȱofȱtheȱingredientsȱinȱaȱtypicalȱ
energyȱ drinkȱ determinedȱ thatȱ caffeineȱ hasȱ theȱ greatestȱ
potentialȱtoȱpresentȱaȱhazard.ȱAȱsingleȱ250ȱmlȱservingȱofȱaȱ
typicalȱenergyȱdrinkȱcontainsȱ80ȱmgȱofȱcaffeine.ȱAlthoughȱ
thisȱamountȱisȱhalfȱtheȱ160ȱmgȱofȱcaffeineȱthatȱaȱsingleȱ250ȱ
mlȱ servingȱ ofȱ coffeeȱ canȱ contain,ȱ itȱ isȱ doubleȱ theȱ 40ȱ mgȱ
presentȱ inȱ aȱ 355ȱ mlȱ servingȱ ofȱ aȱ caffeinatedȱ carbonatedȱ
softȱdrinkȱ(CSD).ȱInȱanyȱcase,ȱtheȱamountȱofȱcaffeineȱinȱaȱ
typicalȱ energyȱ drinkȱ canȱ makeȱ aȱ significantȱ contributionȱ
toȱtheȱtotalȱdietaryȱcaffeineȱintake.ȱȱ
ȱ
Inȱ orderȱ toȱ determineȱ theȱ totalȱ dietaryȱ intakeȱ ofȱ
caffeine,ȱ Healthȱ Canadaȱ accessedȱ dataȱ onȱ theȱ
consumptionȱ ofȱ aȱ wideȱ varietyȱ ofȱ foods,ȱ includingȱ
foodsȱ containingȱ caffeine,ȱ byȱ variousȱ ageȱ groupsȱ
(Statisticsȱ Canada,ȱ Canadianȱ Communityȱ Healthȱ
Surveyȱ Cycleȱ 2.2.,ȱ 2004).ȱ Usingȱ foodȱ intakeȱ dataȱ fromȱ
thisȱ survey,ȱ caffeineȱ consumptionȱ wasȱ determinedȱ
fromȱ allȱ foodȱ sources,ȱ includingȱ coffee,ȱ teaȱ andȱ softȱ
drinks.ȱHowever,ȱintakeȱdataȱspecificȱtoȱenergyȱdrinksȱ
wereȱ notȱ availableȱ becauseȱ thereȱ wasȱ anȱ insufficientȱ
numberȱ ofȱ respondentsȱ whoȱ reportedȱ energyȱ drinkȱ
consumptionȱ inȱ theȱ 2004ȱ survey.ȱ Inȱ theȱ absenceȱ ofȱ
intakeȱ dataȱ specificȱ toȱ energyȱ drinks,ȱ theȱ patternsȱ ofȱ
potentialȱexposureȱtoȱcomponentsȱofȱenergyȱdrinksȱcanȱ
beȱ determinedȱ fromȱ intakeȱ dataȱ forȱ aȱ foodȱ forȱ whichȱ
consumptionȱ canȱ beȱ assumedȱ toȱ beȱ similarȱ toȱ thatȱ ofȱ
energyȱ drinks.ȱ Inȱ theȱ followingȱ exposureȱ assessmentȱ
patterns,ȱ intakeȱ dataȱ forȱ caffeinatedȱ carbonatedȱ softȱȱ
ȱ
www.intechopen.com
ȱ
Medianȱintakesȱȱforȱconsumersȱofȱ
CSDsȱifȱEDsȱsubstitutedȱforȱCSDsȱ
byȱvolumeȱ
(mg/kgȱbw/day)ȱ1ȱ
Theȱ Canadianȱ andȱ internationalȱ consumptionȱ dataȱ forȱ
energyȱ drinksȱ areȱ consideredȱ veryȱ limited.ȱ Inȱ addition,ȱ
internationalȱdataȱthatȱareȱavailableȱfromȱearlyȱ2000ȱmayȱ
notȱreflectȱcurrentȱenergyȱdrinkȱconsumptionȱpatternsȱinȱ
Canadaȱ asȱ theseȱ productsȱ haveȱ becomeȱ muchȱ moreȱ
pervasiveȱinȱtheȱCanadianȱmarketplaceȱinȱrecentȱyears.ȱ
2Ȭ3ȱ
4Ȭ5ȱ
6Ȭ8ȱ
9Ȭ11ȱ
(male)ȱ
9Ȭ11ȱ
(female)ȱ
12Ȭ14ȱ
(male)ȱ
12Ȭ14ȱ
(female)ȱ
15Ȭ16ȱ
(male)ȱ
15Ȭ16ȱ
(female)ȱ
17Ȭ19ȱ
(male)ȱ
17Ȭ19ȱ
(female)ȱ
20+ȱ
(male)ȱ
20+ȱ
(female)ȱ
Pregnant
Medianȱintakeȱforȱconsumersȱofȱ
CSDsȱ(mg/kgȱbw/day)ȱ
5.4ȱȱSummaryȱ
Ageȱinȱ
yearsȱ
(gender)
MedianȱintakesȱȱforȱAllȱpersonsȱ
(consumersȱandȱnonȬconsumersȱ
ofȱCSDs)ȱȱifȱEDsȱsubstitutedȱforȱȱ
CSDsȱbyȱvolumeȱ
(mg/kgȱbw/day)ȱ1ȱ
Basedȱ onȱ theȱ mostȱ recentȱ Canadianȱ Communityȱ Healthȱ
Surveyȱ (CCHS),ȱ 2004,ȱ aȱ determinationȱ ofȱ dailyȱ exposureȱ
levelsȱ isȱ notȱ possibleȱ dueȱ toȱ theȱ limitedȱ reportingȱ ofȱ
energyȱdrinkȱconsumptionȱinȱthatȱsurvey.ȱ
ȱ
ACNielsenȱcarriedȱoutȱsurveysȱfromȱ2006Ȭ2009ȱtoȱprovideȱ
nationalȱ salesȱ andȱ consumerȱ purchaseȱ andȱ demographicȱ
dataȱ forȱ Naturalȱ Healthȱ Productsȱ (NHPs)ȱ ofȱ interest.ȱ
Whileȱ theȱ ACNielsenȱ dataȱ offerȱ anȱ indicationȱ ofȱ theȱ
relativeȱ size,ȱ strengthȱ andȱ demandȱ forȱ energyȱ drinksȱ inȱ
Canada,ȱtheȱdataȱdoȱnotȱprovideȱsufficientȱinformationȱtoȱ
determineȱ dailyȱ consumptionȱ levelsȱ ofȱ energyȱ drinksȱ byȱ
Canadians.ȱȱ
drinksȱ wereȱ usedȱ asȱ surrogateȱ dataȱ forȱ energyȱ drinksȱ
inȱ orderȱ toȱ generateȱ possibleȱ patternsȱ ofȱ anȱ exposureȱ
modelȱthatȱincludesȱenergyȱdrinks.ȱȱ
ȱ
MedianȱintakeȱforȱAllȱpersonsȱ
(consumersȱandȱnonȬconsumersȱ
ofȱCSDs)ȱ
(mg/kgȱbw/day)ȱ
5.3ȱCanadianȱsurveyȱdataȱ
0.06ȱ
0.13ȱ
0.14ȱ
0.15ȱ
0.06ȱ
0.13ȱ
0.14ȱ
0.15ȱ
0.98ȱ
1.18ȱ
1.35ȱ
1.19ȱ
2.62ȱȱ
3.19ȱ
3.41ȱ
2.94ȱ
0.14ȱ
0.14ȱ
0.97ȱ
2.46ȱ
0.18ȱ
0.18ȱ
1.05ȱ
2.81ȱ
0.15ȱ
0.15ȱ
0.86ȱ
2.65ȱ
0.37ȱ
0.47ȱ
1.00ȱ
2.57ȱ
0.17ȱ
0.17ȱ
1.08ȱ
2.89ȱ
0.55ȱ
1.26ȱ
1.17ȱ
2.55ȱ
0.62ȱ
0.95ȱ
0.98ȱ
2.75ȱ
2.23ȱ
2.66ȱ
2.67ȱ
4.19ȱ
2.21ȱ
2.46ȱ
2.70ȱ
4.07ȱ
0.48ȱ
0.53ȱ
1.19ȱ
2.18ȱ
1ȱCaffeinatedȱcarbonatedȱsoftȱdrinksȱwereȱsubstitutedȱwithȱenergyȱdrinksȱ
onȱ aȱ volumeȱ basis.ȱ Theȱ caffeineȱ concentrationȱ ofȱ caffeinatedȱ carbonatedȱ
softȱdrinksȱrangedȱfromȱ73ȱtoȱ140ȱppmȱandȱofȱenergyȱdrinksȱwasȱ320ȱppm.ȱȱ
Tableȱ9.ȱExposureȱmodellingȱofȱtotalȱdietaryȱcaffeineȱintakeȱforȱ(1)ȱ
Allȱ personsȱ (consumersȱ andȱ nonȬconsumersȱ ofȱ caffeinatedȱ
carbonatedȱ softȱ drinksȱ (CSDs)),ȱ (2)ȱ Allȱ persons,ȱ ifȱ allȱ CSDsȱ areȱ
replacedȱbyȱtypicalȱenergyȱdrinksȱ(EDs),ȱ(3)ȱConsumersȱofȱCSDs,ȱ
(4)ȱ Consumersȱ ofȱ CSDsȱ ifȱ allȱ CSDȱ consumptionȱ isȱ replacedȱ byȱ
typicalȱenergyȱdrinksȱȱ
Theȱabsenceȱofȱconsumptionȱmodellingȱforȱenergyȱdrinksȱ
isȱ acknowledged.ȱ Therefore,ȱ theȱ modelȱ belowȱ wouldȱ beȱ
moreȱ aȱ representationȱ ofȱ aȱ worstȱ caseȱ scenario,ȱ asȱ
opposedȱ toȱ aȱ descriptionȱ ofȱ anȱ expectedȱ consumptionȱ
pattern.ȱ Otherȱ factorsȱ suchȱ asȱ taste,ȱ costȱ andȱ theȱ timeȱ
betweenȱ servingsȱ (binging)ȱ mayȱ ultimatelyȱ influenceȱ theȱ
leastȱ conservativeȱ andȱ theȱ worstȱ caseȱ modelsȱ forȱ energyȱ
drinkȱconsumption.ȱ
ȱ
19
0.15ȱ
0.37ȱ
0.17ȱ
0.55ȱ
0.62ȱ
6.0ȱ
2.23ȱ
2.21ȱ
4.6ȱ
0.48ȱ
3.4ȱ
2.4ȱ
6.7ȱ
10.3ȱ
14.6ȱ
17.8ȱ
14.3ȱ
13.0ȱ
4.7ȱ
0.14ȱ
0.18ȱ
0.15ȱ
0.47ȱ
0.17ȱ
1.26ȱ
0.95ȱ
2.66ȱ
2.46ȱ
0.53ȱ
13.2ȱ
9.1ȱ
17.2ȱ
21.9ȱ
21.5ȱ
20.8ȱ
30.9ȱ
27.6ȱ
17.5ȱ
15.2ȱ
5.4ȱ
1ȱRMDIȱisȱtheȱrecommendedȱmaximumȱdailyȱintake.ȱȱ
%ȱexceedingȱRMDIȱifȱCSDsȱ
substitutedȱwithȱEDsȱȱ
0.18ȱ
2.9ȱ
2Ȭ3ȱ
4Ȭ5ȱ
6Ȭ8ȱ
9Ȭ11ȱ
(male)ȱ
9Ȭ11ȱ
(female)ȱ
12Ȭ14ȱ
(male)ȱ
12Ȭ14ȱ
(female)ȱ
15Ȭ16ȱ
(male)ȱ
15Ȭ16ȱ
(female)ȱ
17Ȭ19ȱ
(male)ȱ
17Ȭ19ȱ
(female)ȱ
20+ȱ(male)
20+ȱ
(female)ȱ
Pregnant
MedianȱintakesȱforȱCSDȱconsumersȱȱifȱ
CSDsȱsubstitutedȱwithȱEDsȱ(mg/kgȱ
bw/day)ȱ2ȱ
0.14ȱ
3.8ȱ
4.9ȱ
7.8ȱ
8.2ȱ
%ȱofȱCSDȱconsumersȱexceedingȱRMDIȱ
ȱ
0.06ȱ
0.13ȱ
0.14ȱ
0.15ȱ
MedianȱintakeȱforȱconsumersȱofȱCSDsȱȱ
(mg/kgȱbw/day)ȱ
2.3ȱ
3.5ȱ
3.1ȱ
Ageȱinȱ
yearsȱ
(gender)ȱ
0.98ȱ
1.18ȱ
1.35ȱ
1.19ȱ
3.3ȱ
8.3ȱ
13.5ȱ
14.0ȱ
2.62ȱ3ȱ
3.19ȱ
3.41ȱ
2.94ȱ
56.8ȱ
68.9ȱ
62.9ȱ
61.3ȱ
0.97ȱ
7.3ȱ
2.46ȱ
46.7ȱ
1.05ȱ
8.2ȱ
2.81ȱ
59.3ȱ
0.86ȱ
6.7ȱ
2.65ȱ
55.7ȱ
1.00ȱ
12.6ȱ
2.57ȱ
52.0ȱ
1.08ȱ
20.4ȱ
2.89ȱ
60.6ȱ
1.17ȱ
19.6ȱ
2.55ȱ
59.0ȱ
0.98ȱ
17.3ȱ
2.75ȱ
54.1ȱ
6.0ȱ
2.67ȱ
2.70ȱ
16.2ȱ
15.5ȱ
4.19ȱ
4.07ȱ
29.4ȱ
28.1ȱ
4.6ȱ
1.19ȱ
9.7ȱ
2.18ȱ
14.9ȱ
RMDIȱ1ȱ
(mg/kgȱbw/day)ȱ
0.06ȱ
0.13ȱ
0.14ȱ
0.15ȱ
%ȱofȱAllȱpersonȱexceedingȱRMDIȱifȱ
CSDsȱsubstitutedȱwithȱEDsȱȱ
Medianȱintakesȱforȱ
ȱAllȱpersonsȱifȱCSDsȱsubstitutedȱ
withȱEDsȱ(mg/kgȱbw/day)ȱ2ȱ
2.5ȱ
%ȱofȱAllȱpersonsȱexceedingȱRMDIȱ
2Ȭ3ȱ
4Ȭ5ȱ
6Ȭ8ȱ
9Ȭ11ȱ
(male)ȱ
9Ȭ11ȱ
(female)ȱ
12Ȭ14ȱ
(male)ȱ
12Ȭ14ȱ
(female)ȱ
15Ȭ16ȱ
(male)ȱ
15Ȭ16ȱ
(female)ȱ
17Ȭ19ȱ
(male)ȱ
17Ȭ19ȱ
(female)ȱ
20+ȱ
(male)ȱ
20+ȱ
(female)ȱ
Pregnantȱ
MedianȱintakeȱforȱAllȱpersonsȱforȱ
CSDsȱ
(mg/kgȱbw/day)ȱ
Ageȱinȱ
yearsȱ
(gender)ȱ
RMDIȱ1ȱ
(mg/kgȱbw/day)ȱ
!
2.5ȱ
1ȱ Recommendedȱ Maximumȱ Dailyȱ Intake.ȱ Adolescentsȱ areȱ conservativelyȱ
2ȱCaffeinatedȱcarbonatedȱsoftȱdrinksȱwereȱsubstitutedȱwithȱenergyȱdrinksȱ
assumedȱtoȱhaveȱaȱRMDIȱequalȱtoȱchildren.ȱȱ
2ȱCaffeinatedȱcarbonatedȱsoftȱdrinksȱwereȱsubstitutedȱwithȱenergyȱ drinksȱ
3ȱFiguresȱinȱboldȱequalȱorȱexceedȱtheȱRMDIȱforȱcaffeineȱintakeȱforȱthatȱageȱ
group.ȱ
3ȱFiguresȱinȱboldȱequalȱorȱexceedȱtheȱRMDIȱforȱcaffeineȱintakeȱforȱthatȱageȱ
Tableȱ 10.ȱ Potentialȱ Healthȱ Riskȱ forȱ (1)ȱ Allȱ personsȱ (consumersȱ
andȱnonȬconsumersȱofȱcaffeinatedȱcarbonatedȱsoftȱdrinksȱ(CSDs),ȱ
(2)ȱ Allȱ personsȱ ifȱ allȱ CSDȱ consumptionȱ isȱ replacedȱ byȱ typicalȱ
energyȱdrinksȱ(EDs)ȱ
Tableȱ9ȱshowsȱexposureȱtoȱcaffeineȱforȱallȱpersonsȱandȱforȱ
consumersȱofȱcaffeinatedȱcarbonatedȱsoftȱdrinksȱinȱvariousȱ
ageȱ groups,ȱ expressedȱ asȱ mg/kgȱ bw/day.ȱ Medianȱ caffeineȱ
intakeȱfromȱallȱdietaryȱsourcesȱisȱdepictedȱforȱconsumersȱofȱ
softȱ drinksȱ containingȱ caffeineȱ atȱ “currentȱ marketȱ use”ȱ
levels.ȱTableȱ9ȱalsoȱshowsȱtheȱmedianȱdoseȱofȱcaffeineȱfromȱ
allȱ dietaryȱ sourcesȱ forȱ allȱ personsȱ andȱ consumersȱ ofȱ
caffeinatedȱ carbonatedȱ softȱ drinks,ȱ ifȱ energyȱ drinksȱ wereȱ
substitutedȱ forȱ caffeinatedȱ carbonatedȱ softȱ drinksȱ onȱ aȱ
volumeȱ basisȱ (“substitutedȱ byȱ volume”).ȱ Thisȱ isȱ notȱ toȱ beȱ
construedȱasȱanȱestimateȱofȱexposureȱinȱtheȱtrueȱpopulationȱ
sinceȱ energyȱ drinkȱ consumptionȱ isȱ likelyȱ toȱ beȱ lessȱ onȱ aȱ
volumetricȱscaleȱthanȱcaffeinatedȱcarbonatedȱsoftȱdrinks.ȱ
ȱ
group.ȱ
Tableȱ11.ȱPotentialȱHealthȱRiskȱforȱ(1)ȱConsumersȱofȱcaffeinatedȱ
carbonatedȱ softȱ drinksȱ (CSDs),ȱ (2)ȱ Consumersȱ ofȱ CSDsȱ ifȱ allȱ
caffeinatedȱ carbonatedȱ softȱ drinkȱ consumptionȱ isȱ replacedȱ byȱ
typicalȱenergyȱdrinksȱ(EDs)ȱ
7.ȱPotentialȱHealthȱRiskȱ
Theȱpotentialȱhealthȱriskȱposedȱbyȱtheȱcaffeineȱinȱenergyȱ
drinksȱ canȱ beȱ characterizedȱ byȱ consideringȱ theȱ hazardȱ
dueȱ toȱ caffeineȱ andȱ theȱ potentialȱ exposureȱ toȱ allȱ sourcesȱ
ofȱdietaryȱcaffeine,ȱincludingȱenergyȱdrinks,ȱwithinȱanȱageȱ
groupȱorȱsubgroupȱofȱtheȱpopulation.ȱȱ
ȱ
Healthȱ Canadaȱ usedȱ scientificȱ dataȱ onȱ theȱ healthȱ hazardȱ
posedȱ byȱ caffeineȱ toȱ establishȱ recommendedȱ maximumȱ
dailyȱ intakeȱ (RMDI)ȱ valuesȱ forȱ childrenȱ (2.5ȱ mg/kgȱ
bw/day),ȱ adultsȱ (6.0ȱ mg/kgȱ bw/day)ȱ andȱ womenȱ ofȱ
reproductiveȱ ageȱ (4.6ȱ mg/kgȱ bw/day).ȱ Thereȱ areȱ
www.intechopen.com
!
www.intechopen.com
ȱ
Allȱpersonsȱ
(consumersȱandȱ
nonȬconsumersȱofȱ
CSDs)ȱ
ȱ
%ȱexceedingȱifȱCSDsȱsubstitutedȱ
withȱEDsȱbyȱvolumeȱ
%ȱCSDȱconsumersȱwhoseȱtotalȱ
dietaryȱcaffeineȱintakeȱexceedsȱtheȱ
RMDIȱ
%ȱexceedingȱifȱCSDsȱsubstitutedȱ
withȱEDsȱbyȱvolumeȱ
Percentȱ
consumingȱ
CSDs1ȱ
Consumersȱofȱ
CSDsȱ
%ȱAllȱpersons,ȱincludingȱCSDȱ
consumers,ȱexceedingȱRMDI2ȱ
insufficientȱ dataȱ toȱ determineȱ aȱ separateȱ RMDIȱ forȱ
adolescents.ȱ Asȱ aȱ precautionaryȱ approach,ȱ adolescentsȱ
canȱ beȱ consideredȱ toȱ beȱ asȱ sensitiveȱ asȱ childrenȱ toȱ
caffeine,ȱ inȱ whichȱ case,ȱ theȱ RMDIȱ forȱ childrenȱ canȱ beȱ
conservativelyȱappliedȱtoȱadolescents.ȱȱ
ȱ
Theȱ resultsȱ ofȱ exposureȱ modellingȱ indicateȱ thatȱ theȱ
medianȱ totalȱ dietaryȱ caffeineȱ intakeȱ byȱ allȱ consumersȱ
(Tableȱ 10)ȱ andȱ byȱ consumersȱ ofȱ caffeinatedȱ carbonatedȱ
softȱdrinksȱȱ(Tableȱ11)ȱdoȱnotȱexceedȱtheȱRMDIȱinȱanyȱofȱ
theȱ ageȱ groups.ȱ Theȱ situationȱ forȱ medianȱ intakeȱ remainsȱ
essentiallyȱunchangedȱforȱallȱpersonsȱwhenȱenergyȱdrinksȱ
areȱsubstitutedȱforȱcaffeinatedȱcarbonatedȱsoftȱdrinks,ȱthatȱ
isȱnoȱageȱgroupȱexceedsȱitsȱRMDIȱ(Tableȱ10).ȱInȱcontrast,ȱ
whenȱ energyȱ drinksȱ areȱ substitutedȱ forȱ caffeinatedȱ
carbonatedȱ softȱ drinks,ȱ theȱ medianȱ totalȱ dietaryȱ caffeineȱ
intakeȱ byȱ CSDȱ consumersȱ inȱ everyȱ populationȱ groupȱ ofȱ
childrenȱ andȱ adolescentsȱ meetsȱ orȱ exceedsȱ itsȱ RMDI,ȱ
whileȱthoseȱofȱadultsȱandȱpregnantȱwomenȱdoȱnotȱ(Tableȱ
11).ȱ Theȱ resultsȱ alsoȱ showȱ thatȱ theȱ percentageȱ ofȱ
consumersȱ inȱ eachȱ populationȱ groupȱ whoȱ exceedȱ theirȱ
RMDIȱ increasesȱ dramaticallyȱ whenȱ energyȱ drinksȱ areȱ
substitutedȱforȱcaffeinatedȱcarbonatedȱsoftȱdrinksȱbothȱforȱ
allȱpersonsȱ(Tableȱ10)ȱandȱCSDȱconsumersȱ(Tableȱ11).ȱForȱ
example,ȱ theȱ greatestȱ increaseȱ (fromȱ 3%ȱ toȱ 57%)ȱ inȱ theȱ
percentageȱofȱCSDȱconsumersȱwhoȱexceedȱtheirȱRMDIȱisȱ
observedȱ inȱ 2Ȭȱ toȱ 3ȬyearȬoldsȱ (althoughȱ itȱ isȱ consideredȱ
thatȱ thisȱ scenarioȱ wouldȱ beȱ highlyȱ unlikely,ȱ givenȱ thatȱ
thisȱ ageȱ groupȱ wouldȱ normallyȱ notȱ beȱ givenȱ energyȱ
drinksȱtoȱconsume)ȱ(Tableȱ11).ȱ
ȱ
Usingȱ theȱ dataȱ fromȱ Tablesȱ 10ȱ andȱ 11,ȱ aȱ comparisonȱ forȱ
exposureȱ modellingȱ purposesȱ isȱ madeȱ inȱ Tableȱ 12ȱ ofȱ theȱ
percentageȱ ofȱ allȱ personsȱ andȱ consumersȱ ofȱ CSDsȱ whoȱ
exceedȱtheirȱRMDIȱforȱcaffeineȱunderȱtwoȱscenarios.ȱTheȱ
firstȱ scenarioȱ assumesȱ currentȱ marketȱ use,ȱ whileȱ theȱ
secondȱassumesȱthatȱallȱCSDsȱwereȱreplacedȱwithȱenergyȱ
drinks.ȱ Sinceȱ onlyȱ aȱ portionȱ ofȱ anyȱ givenȱ populationȱ
groupȱ consumesȱ CSDs,ȱ theȱ percentȱ ofȱ CSDȱ consumersȱ
thatȱ exceedsȱ theȱ RMDIȱ forȱ thatȱ populationȱ groupȱ canȱ beȱ
muchȱ greaterȱ thanȱ theȱ percentageȱ ofȱ allȱ personsȱ thatȱ
exceedȱ theȱ RMDI.ȱ Forȱ example,ȱ amongȱ 2Ȭȱ toȱ 3ȬyearȬolds,ȱ
onlyȱ 6%ȱ consumeȱ CSDs.ȱ Replacingȱ CSDsȱ withȱ energyȱ
drinksȱ inȱ thisȱ ageȱ groupȱ bringsȱ theȱ percentageȱ ofȱ
consumerȱ whoȱ exceedȱ theȱ RMDIȱ toȱ 57ȱ %ȱ comparedȱ toȱ
onlyȱ5%ȱofȱallȱpersonsȱwhoȱdoȱso.ȱȱ
ȱ
Theȱ previousȱ sectionȱ presentedȱ aȱ worstȱ caseȱ exposureȱ
scenarioȱinȱwhichȱtheȱconsumptionȱofȱenergyȱdrinksȱwasȱ
patternedȱafterȱthatȱofȱcaffeinatedȱcarbonatedȱsoftȱdrinksȱ
andȱ resultedȱ inȱ moreȱ thanȱ 50%ȱ ofȱ childrenȱ andȱ
adolescents,ȱ aboutȱ 30%ȱ ofȱ adultsȱ andȱ 15%ȱ ofȱ pregnantȱ
femalesȱ exceedingȱ theirȱ respectiveȱ recommendedȱ
maximumȱ dailyȱ intakeȱ (RMDI)ȱ forȱ caffeine.ȱ Itȱ isȱ
importantȱtoȱassessȱhowȱrealisticȱthisȱworstȱcaseȱexposureȱ
scenarioȱis.ȱȱ
2Ȭ3ȱ
6%ȱ
2.3ȱ
4.9ȱ
3.3ȱ
56.8ȱ
4Ȭ5ȱ
7%ȱ
3.5ȱ
7.8ȱ
8.3ȱ
68.9ȱ
6Ȭ8ȱ
12%ȱ
3.1ȱ
8.2ȱ
13.5ȱ
62.9ȱ
9Ȭ11ȱ
(male)ȱ
9Ȭ11ȱ
(female)ȱ
12Ȭ14ȱ
(male)ȱ
12Ȭ14ȱ
(female)ȱ
15Ȭ16ȱ
(male)ȱ
15Ȭ16ȱ
(female)ȱ
17Ȭ19ȱ
(male)ȱ
17Ȭ19ȱ
(female)ȱ
20+ȱ
(male)ȱ
20+ȱ
(female)ȱ
Pregnant
19%ȱ
3.8ȱ
13.2ȱ
14.0ȱ
61.3ȱ
2.9ȱ
9.1ȱ
7.3ȱ
46.7ȱ
3.4ȱ
17.2ȱ
8.2ȱ
59.3ȱ
2.4ȱ
21.9ȱ
6.7ȱ
55.7ȱ
6.7ȱ
21.5ȱ
12.6ȱ
52.0ȱ
10.3ȱ
20.8ȱ
20.4ȱ
60.6ȱ
14.6ȱ
30.9ȱ
19.6ȱ
59.0ȱ
17.8ȱ
27.6ȱ
17.3ȱ
54.1ȱ
14.3ȱ
17.5ȱ
16.2ȱ
29.4ȱ
13.0ȱ
15.2ȱ
15.5ȱ
28.1ȱ
4.7ȱ
5.4ȱ
9.7ȱ
14.9ȱ
Ageȱinȱ
yearsȱ
(gender)
16%ȱ
29%ȱ
22%ȱ
36%ȱ
25%ȱ
40%ȱ
27%ȱ
24%ȱ
16%ȱ
18%ȱ
1ȱ
Dataȱ fromȱ Theȱ Canadianȱ Communityȱ Healthȱ Surveyȱ Bȱ Cycleȱ 2.2ȱ onȱ
Nutrition,ȱStatisticsȱCanada,ȱ2004.ȱȱ
2ȱ RMDIȱisȱtheȱrecommendedȱmaximumȱdailyȱintake.ȱTheȱvalueȱforȱchildren,ȱ
adultsȱandȱpregnantȱfemalesȱareȱ2.5,ȱ6.0ȱandȱ4.6ȱmg/kgȱbw/day,ȱrespectively.ȱ
AdolescentsȱareȱconservativelyȱassumedȱtoȱhaveȱaȱRMDIȱequalȱtoȱchildren.ȱ
Tableȱ12.ȱAȱcomparisonȱofȱpercentȱofȱtotalȱdietaryȱcaffeineȱintakesȱ
thatȱ exceedȱ theȱ caffeineȱ RMDIȱ betweenȱ (1)ȱ consumersȱ ofȱ
caffeinatedȱcarbonatedȱsoftȱdrinksȱ(CSDs),ȱwithȱorȱwithoutȱenergyȱ
drinkȱ substitution;ȱ andȱ (2)ȱ “allȱ persons”ȱ (consumersȱ andȱ nonȬ
consumersȱofȱCSDs),ȱwithȱorȱwithoutȱenergyȱdrinkȱsubstitution.ȱ
Itȱmayȱbeȱreasonablyȱarguedȱthatȱtheȱdietȱofȱchildrenȱ(2Ȭ11ȱ
yearsȱ old)ȱ isȱ monitoredȱ andȱ thatȱ parents,ȱ knowingȱ theȱ
potentialȱ ofȱ adverseȱ effectsȱ ofȱ excessiveȱ caffeine,ȱ areȱ
unlikelyȱtoȱpermitȱtheseȱbeveragesȱinȱtheirȱchildren’sȱdiet.ȱ
Further,ȱ youngȱ childrenȱ (2Ȭ8ȱ yearsȱ old)ȱ doȱ notȱ readilyȱ
haveȱ theȱ abilityȱ toȱ purchaseȱ theseȱ products,ȱ whichȱ areȱ
thereforeȱ muchȱ lessȱ likelyȱ toȱ beȱ partȱ ofȱ theirȱ diet.ȱ Theȱ
percentageȱ ofȱ childrenȱ exceedingȱ theȱ RMDIȱ forȱ caffeineȱ
fromȱtheȱconsumptionȱofȱenergyȱdrinksȱisȱthereforeȱveryȱ
21
!
unlikelyȱ toȱ beȱ inȱ theȱ rangeȱ ofȱ 48ȱ toȱ 69%ȱ asȱ indicatedȱ inȱ
Tableȱ 11ȱ andȱ 12,ȱ whichȱ isȱ basedȱ onȱ theȱ scenarioȱ thatȱ
energyȱ drinksȱ wouldȱ replaceȱ caffeinatedȱ carbonatedȱ softȱ
drinksȱforȱthisȱageȱgroup.ȱȱ
ȱ
Itȱcanȱalsoȱbeȱarguedȱthatȱadultsȱ(20ȱyearsȱandȱolder)ȱareȱ
capableȱ ofȱ monitoringȱ theirȱ ownȱ caffeineȱ intakeȱ andȱ
wouldȱrecognizeȱtheȱacuteȱadverseȱeventsȱassociatedȱwithȱ
excessȱ caffeineȱ intakeȱ andȱ thenȱ moderateȱ theirȱ
consumptionȱ accordingly.ȱ Excessȱ energyȱ drinkȱ
consumptionȱ wouldȱ resultȱ inȱ similarȱ effectsȱ asȱ excessȱ
coffeeȱ consumptionȱ (aȱ typicalȱ energyȱ drinkȱ containsȱ 80ȱ
mgȱofȱcaffeineȱperȱ250ȱmlȱcomparedȱtoȱcupȱofȱcoffeeȱwithȱ
76Ȭ180ȱ mgȱ ofȱ caffeineȱ perȱ 237ȱ ml).ȱ Similarlyȱ pregnantȱ
womenȱ areȱ capableȱ ofȱ monitoringȱ theirȱ ownȱ caffeineȱ
intake.ȱForȱtheseȱreasons,ȱitȱisȱunlikelyȱthatȱtheȱworstȱcaseȱ
exposureȱ scenarioȱ wouldȱ applyȱ toȱ theseȱ groupsȱ withinȱ
theȱgeneralȱpopulation.ȱȱ
ȱ
Adolescentsȱ(12Ȭ19ȱyearsȱold)ȱpresentȱaȱmoreȱcomplicatedȱ
situation.ȱ Inȱ theȱ absenceȱ ofȱ adequateȱ safetyȱ dataȱ toȱ
establishȱ aȱ RMDIȱ ofȱ caffeineȱ forȱ adolescents,ȱ Healthȱ
Canadaȱ followedȱ aȱ precautionaryȱ approachȱ andȱ appliedȱ
theȱ maximumȱ dailyȱ doseȱ recommendedȱ forȱ childrenȱ (2.5ȱ
mg/kgȱ bw/day)ȱ toȱ adolescents.ȱ Theȱ applicationȱ ofȱ theȱ
childrenȇsȱ RMDIȱ toȱ adolescentsȱ isȱ likelyȱ veryȱ
conservative.ȱ Thereȱ isȱ noȱ compellingȱ safetyȱ reasonȱ toȱ
suggestȱ thatȱ 19ȬyearȬoldȱ adolescentsȱ canȱ onlyȱ consumeȱ
2.5ȱ mg/kgȱ bw/dayȱ ofȱ caffeineȱ whereasȱ adultsȱ (20ȱ yearsȱ
andȱolder)ȱcanȱconsumeȱ6ȱmg/kgȱbw/day.ȱItȱisȱmoreȱlikelyȱ
thatȱ adolescentsȱ couldȱ increaseȱ theirȱ maximumȱ dailyȱ
intakeȱ ofȱ caffeineȱ fromȱ 2.5ȱ toȱ 6ȱ mg/kgȱ bw/dayȱ asȱ theyȱ
matureȱwithoutȱappreciableȱhealthȱrisks.ȱNevertheless,ȱinȱ
theȱ absenceȱ ofȱ dataȱ toȱ allowȱ theȱ establishmentȱ ofȱ aȱ
specificȱRMDIȱforȱadolescents,ȱitȱwasȱconsideredȱprudentȱ
toȱ setȱ theȱ maximumȱ recommendedȱ dailyȱ intakeȱ ofȱ
caffeineȱ byȱ adolescentsȱ closerȱ toȱ thatȱ recommendedȱ forȱ
childrenȱthanȱthatȱrecommendedȱforȱadults.ȱȱ
ȱ
Inȱ theȱ worstȱ caseȱ exposureȱ scenario,ȱ describedȱ inȱ theȱ
previousȱ section,ȱ theȱ percentageȱ ofȱ adolescentsȱ whoȱ
exceededȱ theirȱ RMDIȱ rangedȱ fromȱ 47%ȱ toȱ 61%ȱ ofȱ
caffeinatedȱcarbonatedȱsoftȱdrinksȱconsumers,ȱdependingȱ
onȱ ageȱ andȱ genderȱ (Tableȱ 12).ȱ Asȱ suggestedȱ inȱ theȱ
paragraphȱabove,ȱitȱisȱlikelyȱmostȱadolescentsȱcanȱtolerateȱ
greaterȱamountsȱofȱcaffeineȱthanȱtheirȱRMDI.ȱAsȱshownȱinȱ
Tableȱ11,ȱtheȱmedianȱintakesȱofȱcaffeineȱbyȱsubgroupsȱofȱ
adolescentsȱareȱlessȱthanȱ3.0ȱmg/kgȱbw/day,ȱandȱthereforeȱ
onlyȱslightlyȱgreaterȱthanȱtheȱRMDIȱofȱ2.5ȱmg/kgȱbw/day.ȱȱ
ȱ
Theȱ intakeȱ inȱ excessȱ ofȱ theȱ RMDIȱ isȱ similarȱ toȱ whatȱ anȱ
adolescentȱ (12Ȭ19ȱ yearsȱ old)ȱ whoȱ drankȱ aȱ cupȱ ofȱ strongȱ
coffeeȱ (180ȱ mgȱ ofȱ caffeine/237ȱ ml)ȱ wouldȱ beȱ exposedȱ to.ȱ
Forȱ example,ȱ theȱ caffeineȱ intakeȱ fromȱ aȱ singleȱ cupȱ ofȱ
coffeeȱ wouldȱ resultȱ inȱ aȱ doseȱ ofȱ 3.14ȱ andȱ 2.8ȱ mg/kgȱ
bw/dayȱ forȱ anȱ adolescentȱ femaleȱ (aboutȱ 57ȱ kgȱ bw)ȱ andȱ
maleȱ (aboutȱ 64ȱ kgȱ bw)ȱ respectively.ȱ Sinceȱ theseȱ caffeineȱ
ȱ
intakesȱ areȱ veryȱ closeȱ toȱ theȱ conservativeȱ recommendedȱ
maximumȱdailyȱintakeȱofȱ2.5ȱmg/kgȱbw/day,ȱtheyȱwouldȱ
beȱunlikelyȱtoȱposeȱaȱhealthȱhazard.ȱȱ
ȱ
Oneȱ orȱ twoȱ servingsȱ ofȱ aȱ typicalȱ energyȱ drinkȱ (80ȱ mgȱ ofȱ
caffeine/serving)ȱ wouldȱ thereforeȱ beȱ unlikelyȱ toȱ poseȱ anȱ
acuteȱhealthȱhazardȱbasedȱonȱcaffeineȱcontent.ȱHowever,ȱ
itȱisȱknownȱthatȱsomeȱveryȱyoungȱ(12Ȭ14ȱyear)ȱadolescentȱ
malesȱ(aboutȱ56ȱkgȱbw)ȱandȱfemalesȱ(aboutȱ52ȱkgȱbw)ȱwillȱ
consumeȱ aȱ largeȱ volumeȱ ofȱ carbonatedȱ softȱ drinkȱ inȱ aȱ
singleȱ episode,ȱ aboutȱ 760ȱ mlȱ andȱ 519ȱ ml,ȱ respectivelyȱ
(Canadianȱ Communityȱ Healthȱ Surveyȱ 2.2,ȱ 2004).ȱ Thisȱ isȱ
furtherȱcorroboratedȱbyȱtheȱprevalenceȱofȱlargerȱvolumesȱ
ofȱ energyȱ drinkȱ cans,ȱ inȱ comparisonȱ toȱ otherȱ softȱ drinkȱ
cansȱ (710ȱ mlȱ cansȱ versusȱ 355ȱ mlȱ cans).ȱ Ifȱ thisȱ patternȱ ofȱ
consumptionȱisȱappliedȱtoȱenergyȱdrinks,ȱthenȱthereȱcouldȱ
beȱveryȱyoungȱadolescentsȱwhoȱdrinkȱtwoȱcontainersȱofȱaȱ
250ȱmlȱenergyȱdrink,ȱorȱoneȱcontainerȱofȱaȱ473ȱmlȱenergyȱ
drink,ȱandȱyoungȱadolescentȱmalesȱwhoȱcouldȱdrinkȱaȱfullȱ
710ȱ mlȱ energyȱ drinkȱ inȱ aȱ singleȱ episode.ȱ Givenȱ theȱ
amountȱ ofȱ caffeineȱ inȱ typicalȱ andȱ someȱ nonȬtypicalȱ
energyȱ drinks,ȱ theseȱ adolescentsȱ couldȱ haveȱ excessiveȱ
caffeineȱ intake,ȱ especiallyȱ whenȱ consumingȱ largeȱ formatȱ
energyȱdrinks.ȱ
ȱ
Givenȱthatȱmarketingȱforȱenergyȱdrinksȱtendȱtoȱtargetȱsomeȱ
subsetsȱofȱadolescents,ȱthatȱtheyȱareȱcapableȱofȱpurchasingȱ
theseȱproductsȱ(unlikeȱchildren),ȱandȱareȱlessȱfamiliarȱwithȱ
adverseȱ effectsȱ ofȱ excessiveȱ caffeineȱ (unlikeȱ adults),ȱ itȱ isȱ
reasonableȱ toȱ concludeȱ thatȱ theȱ caffeineȱ presentȱ inȱ energyȱ
drinksȱcouldȱposeȱaȱhealthȱriskȱforȱadolescents.ȱ
7.1ȱSummaryȱ
Basedȱ onȱ theȱ hazardȱ characterisationȱ ofȱ eachȱ ofȱ theȱ
ingredientsȱinȱaȱtypicalȱenergyȱdrink,ȱasȱdefinedȱinȱTableȱ
1,ȱitȱwasȱdeterminedȱthatȱtheȱexposureȱassessmentȱshouldȱ
focusȱ onȱ caffeine.ȱ However,ȱ limitedȱ informationȱ isȱ
availableȱ toȱ dateȱ onȱ actualȱ consumptionȱ ofȱ energyȱ drinkȱ
productsȱ inȱ Canada.ȱ Researchȱ isȱ beingȱ consideredȱ toȱ fillȱ
consumptionȱ dataȱ gapsȱ andȱ toȱ provideȱ furtherȱ insightsȱ
intoȱ whetherȱ theȱ labelȱ instructionsȱ relatedȱ toȱ
consumptionȱ adviceȱ onȱ energyȱ drinksȱ areȱ effectiveȱ (i.e.ȱ
areȱfollowed).ȱȱ
ȱ
Inȱ theȱ absenceȱ ofȱ criticalȱ consumptionȱ data,ȱ exposureȱ
modellingȱ wasȱ conductedȱ byȱ substituting,ȱ onȱ aȱ volumeȱ
basis,ȱ energyȱ drinksȱ forȱ caffeinatedȱ carbonatedȱ softȱ drinks,ȱ
forȱwhichȱthereȱisȱintakeȱdata.ȱThisȱapproachȱconservativelyȱ
assumesȱ thatȱ energyȱ drinksȱ areȱ consumedȱ inȱ aȱ mannerȱ
similarȱ toȱ thatȱ forȱ caffeinatedȱ carbonatedȱ softȱ drinks,ȱ andȱ
thatȱenergyȱdrinkȱlabelȱinstructionsȱareȱnotȱfollowed.ȱItȱalsoȱ
assumesȱthatȱaȱvolumeȱbasisȱofȱsubstitutionȱisȱaȱmoreȱlikelyȱ
thanȱaȱservingȱbasisȱofȱsubstitution.ȱȱ
ȱ
Theȱresultsȱofȱtheȱexposureȱassessmentȱthatȱconsideredȱaȱ
worstȱ caseȱ scenarioȱ determinedȱ thatȱ ofȱ theȱ consumersȱ
www.intechopen.com
!
whoȱ replaceȱ caffeinatedȱ carbonatedȱ softȱ drinksȱ withȱ
energyȱ drinks,ȱ moreȱ thanȱ 50%ȱ ofȱ childrenȱ andȱ
adolescents,ȱ aboutȱ 30%ȱ ofȱ adultsȱ andȱ 15%ȱ ofȱ pregnantȱ
femalesȱ wouldȱ exceedȱ theirȱ respectiveȱ recommendedȱ
maximumȱdailyȱintakesȱofȱcaffeineȱandȱthereforeȱmayȱbeȱ
susceptibleȱ toȱ theȱ adverseȱ effectsȱ associatedȱ withȱ excessȱ
caffeineȱ consumption.ȱ Consumersȱ ofȱ caffeinatedȱ
carbonatedȱ softȱ drinksȱ representȱ roughlyȱ 8%ȱ ofȱ youngȱ
childrenȱ(1Ȭȱ8ȱyearsȱold),ȱ22%ȱofȱolderȱchildrenȱ(9Ȭ14ȱyearsȱ
old),ȱ32%ȱofȱadolescents,ȱ20%ȱofȱtheȱadultȱpopulationȱandȱ
13%ȱofȱpregnantȱfemales.ȱȱ
ȱ
Theȱ amountȱ ofȱ caffeineȱ inȱ energyȱ drinksȱ wouldȱ poseȱ aȱ
healthȱ concernȱ ifȱ consumedȱ byȱ childrenȱ (2Ȭ11ȱ yearsȱ old).ȱ
However,ȱ parentsȱ areȱ likelyȱ toȱ beȱ awareȱ thatȱ excessiveȱ
caffeineȱ canȱ beȱ harmfulȱ andȱ wouldȱ keepȱ energyȱ drinksȱ
outȱ ofȱ theirȱ children’sȱ diets.ȱ Energyȱ drinksȱ areȱ thereforeȱ
notȱaȱhealthȱconcernȱforȱthisȱageȱgroup.ȱȱ
ȱ
Adultsȱ (20ȱ yearsȱ andȱ older)ȱ andȱ pregnantȱ womenȱ areȱ
capableȱ ofȱ monitoringȱ theirȱ ownȱ caffeineȱ intake.ȱ Theyȱ
wouldȱ recognizeȱ acuteȱ adverseȱ eventsȱ associatedȱ withȱ
excessȱintakeȱandȱmoderateȱtheirȱconsumptionȱaccordingly.ȱȱ
ȱ
Theȱ situationȱ ofȱ adolescentsȱ (12Ȭ18ȱ yearsȱ old)ȱ isȱ moreȱ
complex.ȱ Becauseȱ ofȱ inadequateȱ safetyȱ dataȱ currentlyȱ
available,ȱ Healthȱ Canadaȱ conservativelyȱ appliedȱ theȱ
RMDIȱforȱchildrenȱ(2.5ȱmg/kgȱbw/day)ȱtoȱthisȱageȱgroupȱ
usingȱaȱworstȱcaseȱscenarioȱ(patterningȱenergyȱdrinkȱuseȱ
onȱ caffeinatedȱ carbonatedȱ softȱ drinkȱ consumption)ȱ toȱ
determineȱ potentialȱ healthȱ risk.ȱ Theȱ percentageȱ ofȱ
adolescentsȱ whoȱ exceededȱ theirȱ RMDIȱ rangedȱ fromȱ 47%ȱ
toȱ 61%ȱ ofȱ caffeinatedȱ carbonatedȱ softȱ drinksȱ consumers.ȱ
But,ȱatȱlessȱthanȱ3.0ȱmg/kgȱbw/day,ȱmedianȱintakesȱwereȱ
onlyȱ slightlyȱ greaterȱ thanȱ theȱ RMDIȱ andȱ likelyȱ toȱ beȱ
tolerated.ȱTheȱcaffeineȱcontentȱofȱoneȱorȱtwoȱservingsȱofȱaȱ
typicalȱ energyȱ drinkȱ (80ȱ mgȱ caffeine/serving)ȱ wouldȱ beȱ
unlikelyȱtoȱposeȱanȱacuteȱhealthȱhazard.ȱ
ȱ
However,ȱ someȱ youngȱ adolescentsȱ (12Ȭ14ȱ year)ȱ consumeȱ
largeȱvolumesȱofȱcaffeinatedȱcarbonatedȱsoftȱdrinksȱ(upȱtoȱ
760ȱ ml)ȱ inȱ aȱ singleȱ episode.ȱ Applyingȱ thisȱ patternȱ ofȱ
consumptionȱ toȱ energyȱ drinksȱ couldȱ resultȱ inȱ excessiveȱ
caffeineȱintakeȱbyȱaȱproportionȱofȱyoungȱadolescents.ȱȱ
ȱ
Inȱconclusion,ȱtheȱmodellingȱexposureȱscenarioȱasȱappliedȱ
toȱ childrenȱ andȱ adultsȱ isȱ unlikelyȱ toȱ representȱ aȱ realisticȱ
healthȱ risk.ȱ However,ȱ forȱ adolescentsȱ theȱ likelihoodȱ ofȱ aȱ
healthȱriskȱisȱgreater,ȱwhichȱmayȱreflectȱtheȱconservativeȱ
assumptionsȱmadeȱaboutȱthisȱsubpopulation.ȱȱ
8.ȱOverallȱSummaryȱandȱConclusionsȱ
Theȱpurposeȱofȱthisȱdocumentȱisȱtoȱprovideȱaȱhealthȱriskȱ
assessmentȱ ofȱ energyȱ drinkȱ productsȱ whenȱ theyȱ areȱ
consumedȱasȱfoodsȱinȱCanada.ȱ
www.intechopen.com
ȱ
Inȱthisȱdocumentȱaȱtypicalȱenergyȱdrinkȱformulationȱwasȱ
definedȱbyȱitsȱingredientsȱandȱservingȱsize,ȱwhereȱaȱsingleȱ
canȱservingȱofȱ250ȱmlȱcontainsȱ80ȱmgȱofȱcaffeine,ȱ1000ȱmgȱ
ofȱ taurine,ȱ 600ȱ mgȱ ofȱ glucuronolactoneȱ andȱ severalȱ Bȱ
vitamins.ȱȱ
ȱ
Theȱliteratureȱonȱhealthȱeffectsȱofȱenergyȱdrinksȱisȱlimitedȱ
toȱlessȱthanȱaȱdozenȱstudies,ȱeachȱwithȱaȱsmallȱnumberȱofȱ
subjectsȱ (N<ȱ 50ȱ subjects/study).ȱ Theseȱ studiesȱ
demonstrateȱ thatȱ consumingȱ 1Ȭ2ȱ servingsȱ ofȱ aȱ typicalȱ
energyȱ drinkȱ formulationȱ willȱ temporarilyȱ increaseȱ
alertnessȱ andȱ attentiveness,ȱ asȱ wellȱ asȱ temporarilyȱ
increaseȱ bloodȱ pressureȱ andȱ heartȱ rate,ȱ inȱ aȱ mannerȱ
similarȱtoȱotherȱcaffeinatedȱbeverages.ȱȱ
ȱ
Studiesȱ thatȱ examineȱ energyȱ drinkȱ useȱ priorȱ toȱ exerciseȱ
showedȱ thatȱ itȱ mayȱ enhanceȱ physicalȱ endurance.ȱ Inȱ
contrast,ȱtheȱconsumptionȱofȱanȱenergyȱdrinkȱafterȱexerciseȱ
mayȱresultȱinȱaȱdelayȱofȱtheȱreturnȱtoȱaȱrestingȱheartȱrate.ȱȱ
ȱ
Studiesȱ alsoȱ suggestȱ thatȱ theȱ combinedȱ consumptionȱ ofȱ
energyȱ drinksȱ andȱ alcoholȱ mayȱ poseȱ aȱ healthȱ risk.ȱ Theȱ
potentialȱ consequenceȱ isȱ thatȱ aȱ personȱ consumingȱ anȱ
energyȱdrinkȱ withȱalcoholȱmayȱbecomeȱintoxicatedȱmoreȱ
quicklyȱ andȱ mayȱ beȱ lessȱ awareȱ ofȱ theirȱ physicalȱ
impairment,ȱ whichȱ mayȱ leadȱtoȱriskyȱ behaviour,ȱ suchȱ asȱ
excessiveȱalcoholȱconsumption.ȱAtȱpresent,ȱclearȱevidenceȱ
supportingȱthisȱsuggestionȱisȱlackingȱbutȱitȱmeritsȱfurtherȱ
investigation.ȱ
ȱ
Itȱwasȱconcludedȱthatȱtheȱpublishedȱinformationȱavailableȱ
onȱ energyȱ drinksȱ wasȱ insufficientȱ toȱ characterizeȱ theȱ
hazardȱ thatȱ thisȱ productȱ asȱ aȱ wholeȱ mayȱ pose.ȱ Forȱ thisȱ
reason,ȱ aȱ reviewȱ ofȱ eachȱ ofȱ theȱ majorȱ ingredientsȱ
containedȱ inȱ aȱ typicalȱ energyȱ drinkȱ wasȱ conducted,ȱ i.e.,ȱ
theȱ potentialȱ healthȱ effectsȱ ofȱ caffeine,ȱ taurine,ȱ
glucuronolactone,ȱ inositolȱ andȱ theȱ Bȱ vitaminsȱ wereȱ
assessed.ȱ Inȱ doingȱ so,ȱ andȱ inȱ contrastȱ withȱ previouslyȱ
publishedȱ information,ȱ theȱ adverseȱ reactionȱ dataȱ relatedȱ
toȱ eachȱ ofȱ theȱ ingredientsȱ inȱ theȱ formulationȱ thatȱ wereȱ
presentedȱ inȱ thisȱ documentȱ attemptedȱ toȱ reflectȱ toȱ theȱ
extentȱ possibleȱ “theȱ realȱ worldȱ use”ȱ ofȱ theȱ productsȱ asȱ
formulated,ȱinȱvariousȱuserȱpopulations.ȱȱ
ȱ
Withȱ respectȱ toȱ caffeine,ȱ anȱ earlierȱ reviewȱ conductedȱ byȱ
HealthȱCanada’sȱFoodȱDirectorateȱconcludedȱthatȱaȱhealthyȱ
adultȱcouldȱtolerateȱaȱmaximumȱintakeȱofȱ400ȱmgȱcaffeineȱ
perȱ day,ȱ whichȱ couldȱ beȱ equivalentȱ toȱ 5ȱ servingsȱ ofȱ aȱ
typicalȱenergyȱdrinkȱperȱday.ȱThisȱamountȱofȱcaffeineȱwasȱ
notȱassociatedȱwithȱadverseȱeffectsȱsuchȱasȱgeneralȱtoxicity,ȱ
cardiovascularȱ effects,ȱ effectsȱ onȱ boneȱ status,ȱ changesȱ inȱ
behaviour,ȱ effectsȱ onȱ maleȱ fertilityȱ orȱ increasedȱ incidenceȱ
ofȱ cancerȱ development.ȱ Further,ȱ theȱ studyȱ concludedȱ thatȱ
reproductiveȬagedȱ womenȱ couldȱ tolerateȱ aȱ maximumȱ
intakeȱ ofȱ 300ȱ mgȱ caffeineȱ perȱ day,ȱ basedȱ onȱ reproductiveȱ
considerations.ȱItȱwasȱalsoȱconcludedȱthatȱchildren,ȱagedȱ4ȱ
23
!
toȱ 12ȱ years,ȱ couldȱ tolerateȱ aȱ maximumȱ ofȱ 45ȱ toȱ 85ȱ mgȱ ofȱ
caffeineȱ perȱ day,ȱ basedȱ onȱ transientȱ mildȱ behaviourȱ
changes.ȱConcerningȱadolescents,ȱagedȱ13ȱtoȱ18ȱyears,ȱthereȱ
wereȱ insufficientȱ dataȱ toȱ determineȱ aȱ dailyȱ maximumȱ
intakeȱ ofȱ caffeine.ȱ However,ȱ theȱ adultȱ recommendedȱ
maximumȱdailyȱintakeȱwasȱconsideredȱinappropriateȱsinceȱ
manyȱ adolescentsȱ haveȱ aȱ lowerȱ bodyȱ weightȱ thanȱ theȱ
averageȱ adult,ȱ butȱ moreȱ importantlyȱ theȱ growingȱ
adolescentȱwasȱconsideredȱtoȱbeȱmoreȱsensitiveȱtoȱcaffeine.ȱ
Dueȱ toȱ thisȱ uncertainty,ȱ itȱ wasȱ moreȱ recentlyȱ concludedȱ
that,ȱ asȱ aȱ precaution,ȱ adolescentsȱ shouldȱ consumeȱ aȱ doseȱ
noȱ greaterȱ thanȱ thatȱ forȱ childrenȱ (2.5ȱ mg/kgȱ bw/day).ȱ Atȱ
thisȱ dose,ȱ adolescentsȱ couldȱ consumeȱ 100ȱ toȱ 175ȱ mgȱ ofȱ
caffeineȱ daily,ȱ dependingȱ onȱ theȱ individualȱ bodyȱ weightȱ
(estimatedȱ toȱ rangeȱ fromȱ 40–70ȱ kg).ȱ LighterȬweightȱ
adolescentsȱ (40ȱ kgȱ bw)ȱ wouldȱ meetȱ theȱ suggestedȱ
maximumȱ intakeȱ ofȱ caffeineȱ withȱ slightlyȱ moreȱ thanȱ aȱ
singleȱ servingȱ ofȱ aȱ typicalȱ energyȱ drink,ȱ whereasȱ heavierȱ
teensȱ (70ȱ kg)ȱ couldȱ consumeȱ 2ȱ servingsȱ perȱ dayȱ inȱ theȱ
absenceȱofȱotherȱsourcesȱofȱdietaryȱcaffeine.ȱȱ
ȱ
Taurine,ȱglucuronolactoneȱandȱinositolȱwereȱconsideredȱtoȱ
beȱ normalȱ constituentsȱ ofȱ theȱ dietȱ andȱ easilyȱ handledȱ byȱ
ordinaryȱ metabolicȱ processes.ȱ Whileȱ generallyȱ consideredȱ
toȱ beȱ ofȱ veryȱ lowȱ orȱ lowȱ toxicity,ȱ thereȱ wasȱ someȱ
uncertaintyȱ aboutȱ theȱ safetyȱ ofȱ theȱ levelsȱ ofȱ theseȱ
substancesȱ inȱ aȱ typicalȱ energyȱ drink,ȱ sinceȱ theȱ levelsȱ ofȱ
someȱ greatlyȱ exceededȱ theȱ amountȱ consumedȱ throughȱ aȱ
normalȱdiet.ȱForȱexample,ȱaȱhighȱdietaryȱintakeȱofȱtaurineȱisȱ
estimatedȱtoȱbeȱ400ȱmg/adult/day,ȱwhereasȱtheȱintakeȱfromȱ
5ȱ servingsȱ ofȱ aȱ typicalȱ energyȱ drinkȱ wouldȱ provideȱ 5000ȱ
mg/adultȱ /day.ȱ Glucuronolactoneȱ hasȱ anȱ estimatedȱ dailyȱ
dietaryȱ intakeȱ ofȱ 2.3ȱ mg/adult/day,ȱ comparedȱ toȱ theȱ 3000ȱ
mg/adult/dayȱ intakeȱ fromȱ 5ȱ servingsȱ ofȱ aȱ typicalȱ energyȱ
drink.ȱ Basedȱ onȱ aȱ reviewȱ ofȱ 90Ȭdayȱ feedingȱ studiesȱ inȱ
laboratoryȱanimals,ȱknowledgeȱofȱtheȱmetabolismȱofȱtheseȱ
twoȱ substancesȱ andȱ informationȱ aboutȱ theirȱ experimentalȱ
therapeuticȱ use,ȱ itȱ wasȱ concludedȱ thatȱ noȱ hazardȱ wasȱ
demonstratedȱatȱtheȱlevelȱofȱ5ȱservingsȱperȱdayȱinȱtheȱshortȬ
term;ȱ however,ȱ longȬtermȱ useȱ inȱ theȱ dietȱ posedȱ anȱ
uncertainty.ȱ Further,ȱ thereȱ wasȱ uncertaintyȱ aboutȱ theȱ
possibleȱinteractionȱofȱtaurineȱwithȱcaffeineȱonȱtheȱnervousȱ
system.ȱȱ
ȱ
Concerningȱ theȱ Bȱ vitamins,ȱ itȱ wasȱ concludedȱ thatȱ theȱ
consumptionȱ ofȱ 5ȱ servingsȱ ofȱ aȱ typicalȱ energyȱ drinkȱ perȱ
dayȱwouldȱnotȱexceedȱtheȱtolerableȱupperȱlimitȱofȱmostȱofȱ
theseȱvitaminsȱandȱtheyȱwouldȱbeȱunlikelyȱtoȱposeȱaȱhealthȱ
riskȱinȱtheȱshortȱterm.ȱThereȱisȱuncertaintyȱinȱtheȱsafetyȱofȱ
lifeȬlongȱconsumptionȱatȱthisȱlevelȱsinceȱtheȱrestȱofȱtheȱdietȱ
wouldȱ alsoȱ contributeȱ toȱ theȱ totalȱ intakeȱ ofȱ theseȱ
substances,ȱ whichȱ couldȱ leadȱ toȱ excessiveȱ intakeȱ andȱ
potentialȱ adverseȱ healthȱ effects.ȱ Niacin,ȱ whichȱ canȱ beȱ
presentȱ asȱ nicotinamideȱ orȱ nicotinicȱ acid,ȱ wasȱ aȱ uniqueȱ
case.ȱ Whenȱ presentȱ asȱ nicotinicȱ acid,ȱ theȱ amountȱ inȱ 1ȱ
servingȱ ofȱ aȱ typicalȱ energyȱ drinkȱ (20ȱ mg)ȱ wasȱ notȱ greaterȱ
ȱ
thanȱtheȱtolerableȱupperȱlimitȱforȱniacinȱ(35ȱmg/adult/day)ȱ
andȱ isȱ unlikelyȱ toȱ poseȱ aȱ healthȱ concern.ȱ However,ȱ theȱ
consumptionȱofȱ2ȱservingsȱperȱdayȱcouldȱpotentiallyȱresultȱ
inȱ flushing.ȱ Inȱ contrast,ȱ whenȱ niacinȱ isȱ presentȱ asȱ
nicotinamide,ȱ thereȱ areȱ noȱ healthȱ concernsȱ atȱ theȱ
consumptionȱ levelȱ ofȱ 5ȱ servingsȱ perȱ day,ȱ sinceȱ someȱ
jurisdictionsȱhaveȱtolerableȱupperȱlimitsȱofȱnicotinamideȱupȱ
toȱ900ȱmg/adult/day.ȱȱ
ȱ
Basedȱonȱcaffeineȱcontentȱalone,ȱitȱcouldȱbeȱsuggestedȱthatȱ
aȱhealthyȱadultȱcouldȱconsumeȱupȱtoȱ5ȱservingsȱofȱaȱtypicalȱ
energyȱ drinkȱ perȱ dayȱ inȱ theȱ absenceȱ ofȱ anyȱ otherȱ dietaryȱ
sourceȱofȱcaffeine.ȱAȱtypicalȱenergyȱdrinkȱcontainsȱ80ȱmgȱofȱ
caffeineȱ andȱ 5ȱ servingsȱ wouldȱ beȱ equalȱ toȱ 400ȱ mg,ȱ theȱ
maximumȱ dailyȱ intakeȱ ofȱ caffeineȱ forȱ anȱ adult.ȱ Fiveȱ
servingsȱ perȱ dayȱ wouldȱ beȱ equalȱ toȱ 5000ȱ mgȱ ofȱ taurine,ȱ
3000ȱ mgȱ ofȱ glucuronolactoneȱ andȱ 300ȱ mgȱ ofȱ inositol.ȱ
Consideredȱ individually,ȱ theseȱ ingredientsȱ consumedȱ atȱ
theseȱlevelsȱwouldȱnotȱbeȱexpectedȱtoȱcauseȱadverseȱeffectsȱ
inȱtheȱshortȬterm,ȱbasedȱonȱanimalȱstudies.ȱHowever,ȱlongȬ
termȱ studiesȱ onȱ taurineȱ andȱ glucuronolactoneȱ haveȱ notȱ
beenȱ conducted.ȱ Inȱ theȱ absenceȱ ofȱ suchȱ studies,ȱ aȱ betterȱ
understandingȱofȱtheȱhumanȱmetabolismȱandȱphysiologicalȱ
impactȱ ofȱ theseȱ highȱ dosesȱ wouldȱ beȱ requiredȱ toȱ increaseȱ
confidenceȱ inȱ theȱ conclusionȱ thatȱ thoseȱ levelsȱ wouldȱ notȱ
likelyȱcauseȱadverseȱeffects.ȱȱ
ȱ
Inȱ conclusion,ȱ consumingȱ 5ȱ servingsȱ perȱ dayȱ forȱ anȱ adult,ȱ
whereȱ aȱ servingȱ isȱ representedȱ byȱ 250ȱ mlȱ ofȱ theȱ typicalȱ
energyȱdrinkȱproductȱformulation,ȱcannotȱbeȱrecommendedȱ
basedȱ onȱ possibleȱ exposureȱ toȱ otherȱ dietaryȱ sourcesȱ ofȱ
caffeine,ȱ uncertaintiesȱ aboutȱ theȱ longȬtermȱ effectsȱ ofȱ someȱ
ingredients,ȱ theȱ potentialȱ interactionȱ betweenȱ ingredientsȱ
andȱ theȱ impactȱ ofȱ chronicȱ consumptionȱ ofȱ highȱ levelsȱ ofȱ
certainȱBȱvitaminsȱ(seeȱtableȱinȱAppendixȱI).ȱȱ
ȱ
Despiteȱ theȱ uncertaintiesȱ aboutȱ possibleȱ interactionsȱ
betweenȱ someȱ ofȱ theȱ ingredients,ȱ 2ȱ servingsȱ ofȱ aȱ typicalȱ
energyȱ drinkȱ perȱ dayȱ wouldȱ notȱ beȱ expectedȱ toȱ poseȱ aȱ
healthȱ riskȱ forȱ theȱ generalȱ adultȱ population.ȱ Thisȱ
conclusionȱ wasȱ basedȱ onȱ theȱ safetyȱ ofȱ theȱ nonȬcaffeineȱ
ingredientsȱ ofȱ energyȱ drinksȱ atȱ thisȱ levelȱ ofȱ consumptionȱ
andȱ theȱ factȱ thatȱ caffeineȱ fromȱ otherȱ dietaryȱ sourcesȱ inȱ
additionȱ toȱ thatȱ inȱ 2ȱ servingsȱ ofȱ energyȱ drinksȱ wouldȱ notȱ
exceedȱ Healthȱ Canada’sȱ recommendedȱ maximumȱ dailyȱ
intakeȱ ofȱ caffeineȱ forȱ theȱ generalȱ adultȱ population.ȱ Theȱ
consumptionȱofȱenergyȱdrinksȱbyȱotherȱsubȬgroupsȱwouldȱ
needȱ toȱ beȱ limitedȱ basedȱ onȱ theȱ respectiveȱ recommendedȱ
maximumȱ dailyȱ intakeȱ ofȱ caffeine.ȱ However,ȱ itȱ canȱ beȱ
notedȱthatȱcurrentȱcaffeinatedȱenergyȱdrinkȱproductȱlabelsȱ
doȱnotȱ recommendȱconsumptionȱ byȱchildren,ȱpregnantȱorȱ
breastfeedingȱwomen,ȱorȱcaffeineȬsensitiveȱpersons.ȱȱ
ȱ
Thisȱ assessmentȱ wouldȱ alsoȱ applyȱ toȱ otherȱ energyȱ drinks,ȱ
describedȱinȱthisȱreport,ȱwhichȱcontainȱmajorȱingredientsȱatȱ
levelsȱ greaterȱ thanȱ thoseȱ inȱ theȱ typicalȱ formulationȱ ofȱ
www.intechopen.com
!
energyȱdrinkȱ products.ȱTheseȱ productsȱ mayȱ beȱconsumedȱ
byȱ adultsȱ atȱ aȱ levelȱ equivalentȱ toȱ 2ȱ servingsȱ ofȱ aȱ typicalȱ
energyȱ drink.ȱ However,ȱ thereȱ isȱ uncertaintyȱ aboutȱ theȱ
safetyȱ ofȱ consumingȱ greaterȱ amountsȱ ofȱ theseȱ otherȱ
formulations.ȱ Itȱ wasȱ alsoȱ notedȱ thatȱ severalȱ productsȱ
containȱ otherȱ bioactiveȱ ingredientsȱ suchȱ asȱ nutrientsȱ (e.g.ȱ
folicȱ acid)ȱ orȱ otherȱ herbalȱ orȱ naturalȱ extractsȱ (e.g.ȱ ginkgoȱ
biloba).ȱ Theseȱ otherȱ formulationsȱ wouldȱ needȱ toȱ beȱ
reviewedȱonȱaȱcaseȬbyȬcaseȱbasis.ȱȱ
ȱ
Exposureȱ dataȱ toȱ energyȱ drinksȱ inȱ Canadaȱ areȱ limited,ȱ inȱ
thatȱ consumptionȱ informationȱ isȱ notȱ reliableȱ basedȱ onȱ
Canadianȱ surveyȱ data.ȱ Inȱ theȱ absenceȱ ofȱ estimatesȱ ofȱ realȱ
consumptionȱdata,ȱtheȱhealthȱriskȱposedȱbyȱenergyȱdrinksȱ
dueȱ toȱ theȱ caffeineȱ contentȱ wasȱ estimatedȱ byȱ modellingȱ
exposureȱusingȱcaffeinatedȱsoftȱdrinkȱconsumptionȱdataȱasȱ
aȱsurrogateȱforȱenergyȱdrinkȱconsumption.ȱȱ
ȱ
Inȱaȱworstȱcaseȱmodellingȱexposureȱscenario,ȱenergyȱdrinksȱ
wereȱsubstitutedȱforȱcaffeinatedȱcarbonatedȱsoftȱdrinksȱonȱ
aȱvolumeȱbasisȱandȱtheȱenergyȱdrinkȱcaffeineȱconcentrationȱ
setȱtoȱ320ȱppmȱforȱmodellingȱpurposes.ȱTheȱresultsȱshowedȱ
thatȱ ofȱ theȱ consumersȱ whoȱ drinkȱ caffeinatedȱ carbonatedȱ
softȱdrinks,ȱmoreȱthanȱ50%ȱofȱtheȱchildrenȱandȱadolescentsȱ
wouldȱ beȱ aboveȱ theȱ recommendedȱ maximumȱ dailyȱ
caffeineȱ intakeȱ ifȱ allȱ carbonatedȱ softȱ drinksȱ wereȱ replacedȱ
byȱaȱtypicalȱenergyȱdrink.ȱFurther,ȱslightlyȱlessȱthanȱ30%ȱofȱ
maleȱandȱfemaleȱadultsȱandȱaboutȱ15%ȱofȱpregnantȱfemalesȱ
wouldȱ exceedȱ theȱ Healthȱ Canada’sȱ recommendedȱ
maximumȱ dailyȱ intakeȱ forȱ caffeine.ȱ Consumersȱ ofȱ
caffeinatedȱ carbonatedȱ softȱ drinksȱ areȱ howeverȱ onlyȱ aȱ
subsetȱ ofȱ theȱ populationȱ andȱ representȱ roughlyȱ 8%ȱ ofȱ
youngȱchildrenȱ(1Ȭȱ8ȱyearsȱold),ȱ22%ȱofȱolderȱchildrenȱ(9Ȭ14ȱ
yearsȱold),ȱ32%ȱofȱadolescents,ȱ20%ȱofȱtheȱadultȱpopulationȱ
andȱ13%ȱofȱpregnantȱfemales.ȱȱ
ȱ
Althoughȱ theȱ intakeȱ modelingȱ showedȱ thatȱ childrenȱ (2Ȭ11ȱ
yearsȱ old)ȱ wouldȱ exceedȱ Healthȱ Canada’sȱ recommendedȱ
levelsȱ forȱ caffeine,ȱ theȱ correspondingȱ healthȱ concernȱ isȱ
limited,ȱ given,ȱ thatȱ childrenȱ areȱ lessȱ likelyȱ toȱ obtainȱ theseȱ
productsȱ onȱ theirȱ ownȱ andȱ thatȱ parentsȱ areȱ expectedȱ toȱ
keepȱenergyȱdrinksȱoutȱofȱtheirȱchildren’sȱdiets.ȱItȱwasȱalsoȱ
concludedȱ thatȱ adultsȱ (20ȱ yearsȱ andȱ older)ȱ andȱ pregnantȱ
womenȱ areȱ capableȱ ofȱ monitoringȱ theirȱ ownȱ caffeineȱ
intake.ȱTheyȱwouldȱalsoȱbeȱmoreȱlikelyȱtoȱrecognizeȱacuteȱ
adverseȱeventsȱassociatedȱwithȱexcessȱintakeȱandȱmoderateȱ
theirȱconsumptionȱaccordingly.ȱȱ
ȱ
Applyingȱ theseȱ hypotheticalȱ consumptionȱ patternsȱ ofȱ
energyȱ drinksȱ toȱ adolescentsȱ hasȱ identifiedȱ aȱ potentialȱ toȱ
exceedȱ theȱ recommendedȱ caffeineȱ intakeȱ byȱ aȱ significantȱ
proportionȱofȱyoungȱadolescents.ȱTheseȱscenariosȱcouldȱnotȱ
beȱexcluded,ȱgivenȱthatȱenergyȱdrinksȱtendȱtoȱbeȱmarketedȱ
toȱ adolescentsȱ whoȱ (unlikeȱ children)ȱ areȱ capableȱ ofȱ
accessingȱtheseȱproducts,ȱincludingȱtheȱlargerȱvolumes,ȱbutȱ
mayȱ beȱ lessȱ likelyȱ thanȱ adultsȱ toȱ adhereȱ toȱ consumptionȱ
recommendations.ȱ Attentionȱ mayȱ thereforeȱ beȱ warrantedȱ
www.intechopen.com
ȱ
asȱtoȱtheȱlevelsȱofȱcaffeineȱpresentȱinȱenergyȱdrinkȱproductsȱ
madeȱ availableȱ forȱ saleȱ inȱ aȱ largeȱ volumeȱ containerȱ (710ȱ
ml),ȱ whichȱ areȱ becomingȱ prevalentȱ andȱ likelyȱ toȱ beȱ
consumedȱbyȱthisȱsubsetȱofȱtheȱpopulation.ȱ
ȱ
Specificȱ riskȱ managementȱ measureȱ toȱ addressȱ potentiallyȱ
highȱcaffeineȱlevelsȱinȱlargerȱvolumeȱenergyȱdrinkȱproductsȱ
wouldȱhelpȱmitigateȱsomeȱrisksȱassociatedȱwithȱexceedingȱ
Healthȱ Canada’sȱ maximumȱ recommendedȱ caffeineȱ intakeȱ
inȱoneȱconsumptionȱsetting.ȱȱ
ȱ
Healthȱ Canada’sȱ proposedȱ riskȱ managementȱ approachȱ forȱ
energyȱ drinksȱ announcedȱ inȱ Octoberȱ 2011ȱ (http://hcȬ
sc.gc.ca/ahcȬasc/media/nrȬcp/_2011/2011Ȭ132Ȭeng.php)ȱ andȱ
updatedȱ inȱ 2012ȱ (http://hcȬsc.gc.ca/fnȬan/legislation/guideȬ
ld/guidanceȬcafȬdrinkȬboissȬtmaȬamtȬeng.php)ȱhelpsȱaddressȱ
aȱ numberȱ ofȱ theseȱ concerns,ȱ throughȱ settingȱ caffeineȱ
concentrationȱ limits,ȱ totalȱ caffeineȱ amountȱ limits,ȱ andȱ
maximumȱ levelsȱ ofȱ vitamins,ȱ mineralsȱ andȱ otherȱ
formulationȱ constituents.ȱ Caffeineȱ andȱ otherȱ nutritionȱ
labellingȱ requirementsȱ wereȱ alsoȱ imposed.ȱ Theseȱ measuresȱ
contributeȱ toȱ mitigatingȱ someȱ ofȱ theȱ risksȱ relatedȱ toȱ theȱ
consumptionȱ/ȱoverconsumptionȱofȱenergyȱdrinkȱproductsȱinȱ
thoseȱareasȱwhereȱinterventionȱisȱpossibleȱbyȱaȱfederalȱfoodȱ
regulatorȱwithinȱtheȱCanadianȱfoodȱregulatoryȱsystem.ȱ
ȱ
Givenȱ theȱ behaviouralȱ aspectsȱ relatedȱ toȱ someȱ ofȱ theȱ
potentialȱ risksȱ Ȭȱ coȬconsumptionȱ withȱ alcohol,ȱ overȱ
exposureȱ toȱ caffeineȱ dueȱ toȱ excessiveȱ andȱ uninformedȱ
consumptionȱ –ȱ itȱ isȱ acknowledgedȱ thatȱ otherȱ areasȱ ofȱ
intervention,ȱ suchȱ asȱ responsibleȱ marketingȱ andȱ
advertising,ȱasȱwellȱasȱeducationȱandȱawareness,ȱwouldȱbeȱ
requiredȱinȱaȱconcertedȱfashion.ȱ
ȱ
Moreȱ researchȱ andȱ surveillanceȱ wouldȱ beȱ requiredȱ toȱ
ascertainȱ theȱ effectivenessȱ ofȱ theȱ variousȱ facetsȱ ofȱ theȱ
proposedȱriskȱmanagementȱapproach.ȱ
ȱ
Variousȱ dataȱ gapsȱ haveȱ beenȱ notedȱ inȱ thisȱ assessment,ȱ inȱ
particular:ȱȱ
ȱ
ȬDataȱthatȱsupportȱtheȱhazardȱcharacterisationȱinȱparticularȱ
asȱ theyȱ pertainȱ toȱ possibleȱ interactionȱ ofȱ theȱ effectsȱ ofȱ theȱ
variousȱ activeȱ ingredientsȱ inȱ energyȱ drinkȱ productȱ
formulationȱ (e.g.ȱ theȱ combinedȱ effectȱ ofȱ taurineȱ andȱ
caffeineȱatȱhighȱconsumptionȱlevels)ȱ
ȱ
ȬȱDataȱthatȱsupportȱanȱimprovedȱcharacterisationȱofȱenergyȱ
drinkȱ consumptionȱ patternsȱ inȱ Canadaȱ byȱ variousȱ
populationȱsubsetsȱ(e.g.ȱolderȱchildrenȱandȱadolescents).ȱȱ
ȱ
Effortsȱareȱcurrentlyȱunderwayȱtoȱinitiateȱresearchȱactivitiesȱ
enablingȱsomeȱaspectsȱofȱtheȱdataȱcollection.ȱThisȱassessmentȱ
willȱthereforeȱbeȱupdatedȱuponȱavailabilityȱofȱnewȱCanadianȱ
dataȱandȱtakingȱintoȱaccountȱanyȱnewȱfindingsȱonȱtheȱsafetyȱ
ofȱenergyȱdrinksȱdomesticallyȱandȱinternationally.ȱ
25
!
9.ȱAppendixȱ1ȱ
Notȱ
applicableȱ
Notȱ
applicableȱ
160ȱmgȱ
80ȱmgȱ
160ȱmgȱ
400ȱmgȱ
Upperȱtolerableȱ
limitȱ
(UL)/Maximumȱ
recommendedȱ
intakeȱ
(MRI)/Maximumȱ
Limitȱ(ML)ȱ
400ȱmgȱ=ȱMRIȱ
40ȱ–ȱ400ȱ
mgȱ
1000ȱmgȱ
2000ȱmgȱ
5000ȱmgȱ
Notȱestablishedȱ
Glucuronolactoneȱ
Notȱ
applicableȱ
1.2Ȭȱ2.4ȱ
mgȱ
600ȱmgȱ
1200ȱmgȱ
3000ȱmgȱ
Notȱestablishedȱ
Inositolȱ
Notȱ
500Ȭ1000ȱ
applicableȱ
mgȱ
50ȱmgȱ
100ȱmgȱ
250ȱmgȱ
Notȱestablishedȱ
Thiamineȱ
1.1Ȭ1.2ȱmgȱ
4ȱmgȱ
5ȱmgȱ
10ȱmgȱ
25ȱmgȱ
40ȱmgȱ=ȱMLȱ
Riboflavinȱ
1.1Ȭ1.3ȱmgȱ
4.5ȱmgȱ
1.65ȱmgȱ
3.30ȱmgȱ
8.25ȱmgȱ
20ȱmgȱ=ȱULȱ
16ȱmgȱ
77ȱmgȱ
18ȱmgȱ
36ȱmgȱ
90ȱmgȱ
900ȱmgȱ=ȱULȱ
VitaminȱB6ȱ
1.3Ȭ1.7ȱmgȱ
2ȱmgȱ
4ȱmgȱ
10ȱmgȱ
Notȱestablishedȱ
VitaminȱB12ȱ
2.4ȱmcgȱ
0.9Ȭ4.5ȱ
mgȱ
2Ȭ6ȱmcgȱ
1ȱmcgȱ
2ȱmcgȱ
5ȱmcgȱ
20ȱmcgȱ=ȱMLȱ
5ȱmgȱ
3Ȭ12ȱmgȱ
6ȱmgȱ
12ȱmgȱ
30ȱmgȱ
20ȱmgȱ=ȱMLȱ
Dailyȱ
Requiredȱ
dietaryȱ
dailyȱintakeȱ
intakeȱ
Ingredientȱ
Caffeineȱ
Taurineȱ
Niacinȱ
nicotinamide)ȱ
PantothenicȱAcidȱ
(asȱ
Amountȱ
perȱ
servingȱ
Amountȱ
perȱ2ȱ
servingsȱ
Amountȱ
perȱ5ȱ
servingsȱ
Otherȱsafetyȱdataȱ
ȱ
6000ȱmg/dayȱ=ȱNOAELȱ
inȱ42ȱdayȱhumanȱstudyȱ
ȱ
1000ȱmg/kgȱbw/dayȱ=ȱ
NOAELȱinȱ90Ȭdayȱratȱ
studyȱ
asȱexperimentalȱtherapyȱ
inȱhumanȱ
1000ȱmg/kgȱbw/day=ȱ
NOAELȱinȱ90Ȭdayȱratȱ
studyȱ
inȱ42ȱdayȱtherapyȱinȱ
humanȱstudyȱ
NOAELȱinȱ24ȱweekȱ
mouseȱstudyȱ
Littleȱdangerȱofȱthiamineȱ
toxicityȱassociatedȱwithȱ
oralȱintakeȱofȱlargeȱ
amountsȱ(500ȱmgȱdailyȱ
forȱ1ȱmonth)ȱ
NOAELȱinȱ13ȱweekȱ
feedingȱstudyȱinȱrats.ȱ
Noȱpublishedȱdataȱwithȱ
toxicȱeffectsȱinȱhumans.ȱ
3Ȭ9ȱg/dayȱnicotinamideȱ
forȱseveralȱdaysȱresultedȱ
inȱnauseaȱinȱsingleȱ
subjectȱandȱhepatitisȱinȱ
anotherȱsubject.ȱ
100ȱȬ500ȱmgȱcausesȱ
neurologicalȱsymptomsȱ
Oralȱandȱparenteralȱ
supplementationȱwithȱ
dosagesȱbetweenȱ1Ȭ5ȱmgȱ
everyȱ2ȱweeksȱorȱmonthȱ
haveȱbeenȱgivenȱforȱupȱ
toȱatȱleastȱ5ȱyears,ȱinȱ
patientsȱwithȱ
compromisedȱvitaminȱ
B12ȱabsorption,ȱwithoutȱ
anyȱidentifiedȱadverseȱ
effects.ȱ
10000ȱmgȱcausesȱ
gastrointestinalȱeffectsȱinȱ
humansȱ
Notes:ȱMLsȱforȱthiamine,ȱvitaminȱB12ȱandȱpantothenicȱacidȱwereȱspecificallyȱassignedȱtoȱenergyȱdrinksȱbyȱNHPD.ȱTheseȱfiguresȱareȱnotȱULsȱorȱMLsȱ
forȱtheseȱBȱvitaminsȱforȱtheȱtotalȱdiet.ȱULȱforȱniacinȱwasȱdeterminedȱbyȱtheȱEuropeanȱCommunity.ȱȱ
TableȱA1.ȱSummaryȱofȱSafetyȱAssessmentsȱofȱtheȱIngredientsȱinȱaȱTypicalȱEnergyȱDrink
ȱ
www.intechopen.com
!
[15] Canadianȱ Communityȱ Healthȱ Surveyȱ 2004.ȱ
http://www.hcȬsc.gc.ca/fnȬan/surveill/nutrition/ȱ
[1] ȱACNielsen,ȱ Theȱ Nielsenȱ Company,ȱ Naturalȱ Healthȱ
commun/indexȬeng.phpȱ
Productsȱ Retailȱ Salesȱ andȱ Consumerȱ Insightsȱ
[16] CandowȱDG,ȱKleisingerȱAK,ȱGrenierȱS,ȱetȱal.ȱEffectȱofȱ
Summaryȱ Report,ȱ preparedȱ forȱ theȱ Marketedȱ
sugarȬfreeȱ Redȱ Bullȱ Energyȱ Drinkȱ onȱ highȬintensityȱ
Biologicals,ȱ Biotechnologyȱ andȱ Naturalȱ Healthȱ
runȱ timeȬtoȬexhaustionȱ inȱ youngȱ adults.ȱ Jȱ Strengthȱ
Productsȱ Bureau,ȱ Healthȱ Canada,ȱ Marchȱ 2009.ȱȱȱ
CondȱResȱ2009;ȱ23(4):1271Ȭ1275.ȱ
Agricultureȱ andȱ AgriȬFoodȱ Canadaȱ Januaryȱ 2008.ȱ
[17] Chouȱ T.ȱ Wakeȱ upȱ andȱ smellȱ theȱ coffee.ȱ Caffeine,ȱ
Theȱ Energyȱ Drinkȱ Segmentȱ inȱ Northȱ America.ȱ
coffee,ȱ andȱ theȱ medicalȱ consequence.ȱ Westȱ Jȱ Medȱ
1992;ȱ157(5):544Ȭ553.ȱ
http://www.atsȬsea.agr.gc.ca/info/4387Ȭeng.htmȱ
[18] Colodnyȱ Lȱ andȱ Hoffmanȱ RL.ȱ InositolȬclinicalȱ
[2] ȱAgricultureȱ andȱ AgriȬFoodȱ Canadaȱ Aprilȱ 2008.ȱ
applicationsȱ forȱ exogenousȱ use.ȱ Alternȱ Medȱ Revȱ
Marketȱ Brief:ȱ Theȱ Canadianȱ Energyȱ Drinkȱ Market.ȱ
1998;ȱ3(6):432Ȭ447.ȱ
http://www.atsȬsea.agr.gc.ca/can/4469Ȭeng.htmȱ
[19] CommitteeȱonȱToxicityȱofȱChemicalsȱinȱFood,ȱConsumerȱ
[3] ȱAgricultureȱ andȱ AgriȬFoodȱ Canadaȱ Aprilȱ 2011.ȱ
Productsȱ andȱ theȱ Environment.ȱ COTȱ statementȱ onȱ theȱ
Marketȱ Brief:ȱ Theȱ Canadianȱ Energyȱ Drinkȱ Market.ȱ
interactionȱ ofȱ caffeineȱ andȱ alcoholȱ andȱ theirȱ combinedȱ
http://www.atsȬsea.agr.gc.ca/can/4469Ȭeng.htmȱȱ
effectsȱonȱhealthȱandȱbehaviour.ȱ2012.ȱȱ
[4] ȱAhrensȱ DRA,ȱ Douglassȱ DLW,ȱ Flynnȱ MSSM,ȱ etȱ al.ȱ
[20] CossiȱR,ȱRicordyȱR,ȱBartoliniȱF,ȱetȱal.ȱTaurineȱandȱellagicȱ
Lackȱ ofȱ effectȱ ofȱ dietaryȱ supplementsȱ ofȱ glucuronicȱ
acid:ȱ twoȱ differentlyȬactingȱ naturalȱ antioxidants.ȱ
acidȱ andȱ glucuronolactoneȱ onȱ longevityȱ ofȱ theȱ rat.ȱ
EnvironȱMolȱMutagenȱ1995;ȱ26(3):248Ȭ254.ȱ
NutritionȱResearchȱ1987;ȱ7(6):683Ȭ688.ȱ
[21]
Clausonȱ
KA,ȱ Shieldsȱ KM,ȱ McQueenȱ CEȱ andȱ Persadȱ
[5] ȱAlfordȱ C,ȱ Coxȱ Hȱ andȱ Wescottȱ R.ȱ Theȱ effectsȱ ofȱ Redȱ
N.ȱ
Safetyȱ
issuesȱ associatedȱ withȱ commerciallyȱ
Bullȱ Energyȱ Drinkȱ onȱ humanȱ performanceȱ andȱ
availableȱ
energyȱ
drinks.ȱ J.ȱ AmericanPharmacistsȱ
mood.ȱAminoȱAcidsȱ2001;ȱ21(2):139Ȭ150.ȱ
Associationȱ2008;ȱ48(3):ȱe55Ȭe63ȱ
[6] ȱAlfordȱ C,ȱ HamiltonȬMorrisȱ J.,ȱ andȱ Wersterȱ JC.ȱ Theȱ
[22] CurryȱKȱandȱStasioȱMJ.ȱTheȱeffectsȱofȱ‘energyȱdrinks’ȱ
effectsȱ ofȱ energyȱ drinkȱ inȱ combinationȱ withȱ alcoholȱ
aloneȱ andȱ withȱ alcoholȱ onȱ neuropsychologicalȱ
onȱ performanceȱ andȱ subjectiveȱ awareness.ȱ
functioning.ȱ Humȱ Psychopharmacolȱ Clinȱ Expȱ 2009;ȱ
Pscyhopharmacologyȱ2012:ȱ222ȱ(3);ȱ519Ȭ532.ȱ
24(6):473Ȭ481.ȱ
[7] ȱAmericanȱ Heartȱ Association,ȱ 2007.ȱ Energyȱ drinksȱ
[23] Daltonȱ Kȱ andȱ Daltonȱ MJT.ȱ Characteristicsȱ ofȱ
mayȱposeȱrisksȱforȱpeopleȱwithȱhighȱbloodȱpressure,ȱ
pyridoxineȱ overdoseȱ neuropathyȱ syndrome.ȱ Actaȱ
heartȱ disease.ȱ Newsȱ Releaseȱ 11.06.07,ȱ Sicnetificȱ
NeurologicaȱScandinavicaȱ1987;ȱ76:8Ȭ11.ȱ
Sessions,ȱOrlando,ȱFlorida.ȱ
[24] DiamondȱS,ȱBalmȱTKȱandȱFreitagȱFG.ȱIbuprofenȱplusȱ
[8] ȱArendrupȱK,ȱGregersenȱG,ȱHawleyȱJ,ȱetȱal.ȱHighȱdoseȱ
caffeineȱ inȱ theȱ treatmentȱ ofȱ tensionȬtypeȱ headache.ȱ
dietaryȱ myoȬinositolȱ supplementationȱ doesȱ notȱ alterȱ
ClinȱPharmacolȱTherȱ2000;ȱ68(3):312Ȭ319.ȱ
theȱischaemicȱphenomenonȱinȱhumanȱdiabetics.ȱActaȱ
[25] Dohertyȱ M,ȱ andȱ Smithȱ PM.ȱ Effectsȱ ofȱ caffeineȱ
NeurolȱScanȱ1989;ȱ80(2):99Ȭ102.ȱ
ingestionȱ onȱ exerciseȱ testing:ȱ aȱ metaȬanalysis.ȱ Intnȱ Jȱ
[9] ȱAstorino,ȱT.A.ȱandȱRobertson,ȱD.W.,ȱEfficacyȱofȱacuteȱ
SportȱNutrȱExercȱMetabȱ2004;ȱ14:ȱ626Ȭ646.ȱȱ
caffeineȱ ingestionȱ forȱ shortȬtermȱ highȬintensityȱ
[26] Estensenȱ RDȱ andȱ Wattenbergȱ LW.ȱ Studiesȱ ofȱ
exerciseȱperformance:ȱaȱsystematicȱreview.ȱJȱStrengthȱ
chemopreventiveȱ effectsȱ ofȱ myoȬinositolȱ onȱ
CondȱRes.ȱ2010;ȱ24ȱ(1):ȱ257Ȭ265.ȱ
benzo[a]pyreneȬinducedȱ neoplasiaȱ ofȱ theȱ lungȱ andȱ
[10] Australiaȱ Newȱ Zealandȱ Foodȱ Authority:ȱ Inquiryȱ
forestomachȱofȱfemaleȱA/Jȱmice.ȱCarcinogenesisȱ1993;ȱ
Reportȱ Applicationȱ A394,ȱ Formulatedȱ Caffeinatedȱ
14(9):1975Ȭ1977.ȱ
Beveragesȱ8ȱAugustȱ2001.ȱ
[27] Europeanȱ Commission:ȱ Opinionȱ ofȱ theȱ Scientificȱ
http://www.foodstandards.gov.au/foodstandards/ap
Committeeȱ onȱ Foodȱ onȱ theȱ Tolerableȱ Upperȱ Intakeȱ
plications/applicationa394formulatedcaffeinatedbeve
LevelsȱofȱPantothenicȱAcidȱ(expressedȱonȱ17ȱAprilȱ2002).ȱ
ragesenergydrinks/applicationa394formu866.cfm.ȱ
ȱȱȱȱȱȱȱȱhttp://ec.europa.eu/food/fs/sc/scf/out80k_en.pdfȱ
[11] BfRȱ Informationȱ No.016/2008,ȱ 13ȱ Marchȱ 2008.ȱ Newȱȱȱȱȱȱȱȱ [28] Europeanȱ Commission:ȱ Opinionȱ ofȱ theȱ Scientificȱ
HumanȱDataȱonȱtheȱAssessmentȱofȱEnergyȱDrinks.ȱȱ
[12] Bichlerȱ A.,ȱ Swensonȱ A.ȱ andȱ Harrisȱ M.A.ȱ Aȱ
Levelsȱ ofȱ Vitaminȱ B1ȱ (expressedȱ onȱ 11ȱ Julyȱ 2001).ȱ
combinationȱ ofȱ caffeineȱ andȱtaurineȱ hasȱ noȱeffectȱonȱ
http://ec.europa.eu/food/fs/sc/scf/out93_en.pdfȱ
shortȱtermȱmemoryȱbutȱinducesȱchangesȱinȱheartȱrateȱ
andȱmeanȱarterialȱbloodȱpressure.ȱAminoȱAcidsȱ2006;ȱ
31:ȱ471Ȭ476.ȱ
LevelsȱofȱVitaminȱB2ȱ(expressedȱonȱ22ȱNovemberȱ2000).ȱ
[13] BoucknoogheȱT.,ȱRemacleȱC.,ȱReusensȱB.ȱIsȱtaurineȱaȱ
http://ec.europa.eu/food/fs/sc/scf/out80i_en.pdfȱ
functionalȱnutrient?ȱCurrȱOpinȱclinȱNutrȱMetabȱCareȱ
2006;ȱ9:ȱ728Ȭ733.ȱ
LevelsȱofȱVitaminȱB6ȱ(expressedȱonȱ19ȱOctoberȱ2000).ȱ
[14] Burkeȱ LM.ȱ Caffeineȱ andȱ sportsȱ performance.ȱ Applȱ
http://ec.europa.eu/food/fs/sc/scf/out80c_en.pdfȱȱ
PhysiolȱNutrȱMetabȱ2008;ȱ33(6):1319Ȭ1334.ȱ
10.ȱReferencesȱ
www.intechopen.com
ȱ
27
!
LevelsȱofȱVitaminȱB12ȱ(expressedȱonȱ19ȱOctoberȱ2000).ȱ
http://ec.europa.eu/food/fs/sc/scf/out80d_en.pdfȱ
[32] Europeanȱ Foodȱ Safetyȱ Authority,ȱ 15ȱ Januaryȱ 2009.ȱ
Theȱ useȱ ofȱ taurineȱ andȱ DglucuronoȬ(Ȭlactoneȱ asȱ
constituentsȱ ofȱ theȱ soȬcalledȱ “energy”ȱ drinks.ȱ
Scientificȱ Opinionȱ ofȱ theȱ Panelȱ onȱ Foodȱ Additivesȱ
andȱNutrientȱSourcesȱaddedȱtoȱFood.ȱȱ
http://www.efsa.europa.eu/EFSA/efsa_localeȬ
1178620753812_1211902327742.htmȱ
[33] Evansȱ GH,ȱ Shirreffsȱ SMȱ andȱ Maughanȱ RJ.ȱ
Postexerciseȱ rehydrationȱ inȱ man:ȱ theȱ effectsȱ ofȱ
osmolalityȱ andȱ carbohydrateȱ contentȱ ofȱ ingestedȱ
drinks.ȱNutritionȱ2009;ȱ25(9):905Ȭ13.ȱȱ
[34] FerreiraȱSE,ȱdeȱMelloȱMT,ȱPompéiaȱSȱetȱal.ȱEffectsȱofȱ
Energyȱ Drinkȱ Ingestionȱ onȱ Alcoholȱ Intoxication.ȱ
AlcoholȱClinȱExpȱResȱ2006;ȱ30(4):598Ȭ605.ȱ
[35] Foodȱ Safetyȱ Promotionȱ Boardȱ (FSPB).ȱ Aȱ Reviewȱ ofȱ
theȱHealthȱEffectsȱofȱStimulantȱDrinks,ȱ2003.ȱIreland.ȱ
http://www.safefood.eu/Global/Publications/Researc
h%20reports/FSPB%20Stimulant%20drinks%20.pdf?e
pslanguage=enȱ
Foodȱ Standardsȱ Agency.ȱ Expertȱ Groupȱ onȱ Vitaminsȱ
andȱ Minerals:ȱ Safeȱ Upperȱ Levelsȱ forȱ Vitaminsȱ andȱ
Mineralsȱ (Mayȱ 2003).ȱ http://www.food.gov.uk/ȱ
multimedia/pdfs/vitmin2003.pdfȱ
[36] ForbesȱSC,ȱCandowȱDG,ȱLittleȱJPȱetȱal.ȱEffectȱofȱRedȱ
Bullȱ energyȱ drinkȱ onȱ repeatedȱ Wingateȱ cycleȱ
performanceȱ andȱ benchȬpressȱ muscleȱ endurance.ȱ Intȱ
JȱSportȱNutrȱExercȱMetabȱ2007;ȱ17(5):433Ȭ444.ȱ
[37] Garrettȱ BEȱ andȱ Griffithsȱ RR.ȱ Theȱ roleȱ ofȱ dopamineȱ inȱ
theȱ behaviouralȱ effectsȱ ofȱ caffeineȱ inȱ animalsȱ andȱ
humans.ȱPharmacolȱBiochemȱBehavȱ1997;ȱ57(3):533Ȭ541.ȱ
[38] Grahamȱ TE.ȱ Caffeineȱ andȱ Exercise:ȱ Metabolism,ȱ
Enduranceȱ andȱ Performance.ȱ Sportsȱ Medȱ 2001;ȱ
31(11):785Ȭ807.ȱ
[39] Groffȱ JLȱ andȱ Gropperȱ SS.ȱ Advancedȱ Nutritionȱ andȱ
Humanȱ Metabolism,ȱ 3rdȱ Edition.ȱ Wadsworthȱ
ThomsonȱLearningȱ2000.ȱ
[40] HayesȱKCȱandȱTrautweinȱEA,ȱ1994.ȱModernȱnutritionȱ
inȱhealthȱandȱdisease.ȱIn:ȱTaurine,ȱLeaȱ&ȱFebiger.ȱPpȱ
477Ȭ485.ȱ
[41] Healthȱ Canada.ȱ Eatingȱ Wellȱ withȱ Canada’sȱ Foodȱ
Guide.ȱ 2007.ȱ http://www.hcȬsc.gc.ca/fnȬan/pubs/resȬ
educat/resȬeducat_3Ȭeng.php#a6ȱ
[42] Healthȱ Canada,ȱ Draftȱ Reportȱ –ȱ Estimatedȱ Dietaryȱ
ExposuresȱtoȱCaffeineȱforȱCanada,ȱ2010.ȱ
[43] Healthȱ Canada.ȱ It’sȱ Yourȱ Healthȱ –ȱ Obesity.ȱ 2010.ȱ
http://www.hcȬsc.gc.ca/hlȬvs/iyhȬvsv/lifeȬvie/ȱ
obesȬeng.phpȱȱ
[44] HealthȱCanada.ȱIt’sȱYourȱHealthȱ–ȱSafeȱUseȱofȱEnergyȱ
Drinks.ȱ
2010.ȱ
http://www.hcȬsc.gc.ca/hlȬvs/iyhȬ
vsv/foodȬaliment/boissonsȬenergȬdrinksȬeng.phpȱȱ
[45] Higdon,ȱ J.V.ȱ andȱ Frei,ȱ B.ȱ Coffeeȱ andȱ Health:ȱ Aȱ
ReviewȱofȱRecentȱHumanȱResearch.ȱCriticalȱReviewsȱ
inȱFoodȱScienceȱandȱNutrition.ȱ2006.ȱ46:101Ȭ123.ȱ
ȱ
[46] Horneȱ JAȱ andȱ Reynerȱ LA.ȱ Beneficialȱ effectsȱ ofȱ anȱ
“energyȱdrink”ȱgivenȱtoȱsleepyȱdrivers.ȱAminoȱAcidsȱ
2001;ȱ20(1):83Ȭ89.ȱ
[47] Instituteȱ ofȱ Medicineȱ (IOM):ȱ Dietaryȱ Referencesȱ
Intakesȱ forȱ Thiamin,ȱ Riboflavin,ȱ Niacin,ȱ Vitaminȱ B6,ȱ
Folate,ȱ Vitaminȱ B12,ȱ Pantothenicȱ Acid,ȱ Biotinȱ andȱ
Choline.ȱ1998.ȱȱ
(http://www.nap.edu/catalog.php?record_id=6015)ȱ
[48] IvyȱJL,ȱKammerȱL,ȱDingȱZ,ȱetȱal.ȱImprovedȱcyclingȱtimeȬ
trialȱ performanceȱ afterȱ ingestionȱ ofȱ aȱ caffeineȱ energyȱ
drink.ȱIntȱJȱSportȱNutrȱExercȱMetabȱ2009(1);ȱ19:61Ȭ78.ȱ
[49] Kohlerȱ Vȱ andȱ Schmidȱ W.ȱ Dieȱ Behandlungȱ gesunderȱ
SalmonellenȬDauerausscheiderȱ(Enteritisȱthphiȱsalm.)ȱ
mitȱ GlukuonsaureȬgammȬLkton.ȱ Therapieȱ wochȱ
1980;ȱ30:2831Ȭ2833.ȱ
[50] KoponenȱH,ȱLepolaȱU,ȱLeinonenȱE,ȱetȱal.ȱCitalopramȱ
inȱtheȱtreatmentȱofȱobsessiveȬcompulsiveȱdisorder;ȱanȱ
openȱpilotȱstudy.ȱActȱPsychiatȱScandȱ1997;ȱ96(5):ȱ343Ȭ
346.ȱȱ
[51] Kurodaȱ M,ȱ Yoshidaȱ Dȱ andȱ Mizusakiȱ S.ȱ BioȬantiȱ
mutagenicȱ effectȱ ofȱ lactonesȱ onȱ chemicalȱ
mutagenesis.ȱAgricȱBiolȱChemȱ1986;ȱ50:243Ȭ245.ȱȱ
[52] Laidlawȱ SA,ȱ Dietrichȱ MF,ȱ Lamtenzanȱ MPȱ etȱ al.ȱ
Antimutagenicȱ effectsȱ ofȱ taurineȱ inȱ aȱ bacterialȱ assayȱ
system.ȱCancerȱResȱ1989;ȱ49(23):6600Ȭ6604.ȱ
[53] Lockwoodȱ CM,ȱ Moonȱ JR,ȱ Smithȱ AE,ȱ etȱ al.ȱ LowȬ
calorieȱ energyȱ drinkȱ improvesȱ physiologicalȱ
responseȱ toȱ exerciseȱ inȱ previouslyȱ sedentaryȱ men:ȱ aȱ
placeboȬcontrolledȱ efficacyȱ andȱ safetyȱ study.ȱ Jȱ
StrengthȱCondȱResȱ2009ȱ(epub)ȱ
[54] Loristȱ M,ȱ andȱ Topsȱ MM.ȱ Caffeine,ȱ fatigueȱ andȱ
cognition.ȱBrainȱCogn.ȱ2003;ȱ53:82Ȭ94.ȱ
[55] Malinauskasȱ BM,ȱ Aebyȱ VG,ȱ Overtonȱ RF,ȱ etȱ al.ȱ Aȱ
surveyȱofȱenergyȱdrinkȱconsumptionȱpatternsȱamongȱ
collegeȱstudents.ȱNutrȱJȱ2007;ȱ6:35.ȱ
[56] Maughan,ȱ R.J.ȱ andȱ Griffin,ȱ J.ȱ Caffeineȱ ingestionȱ andȱ
fluidȱ balance:ȱ aȱ review.ȱ Jȱ Humȱ Nutrȱ Dieitetȱ 2003;ȱ 16:ȱ
411Ȭ420.ȱȱ
[57] NawrotȱP,ȱJordanȱS,ȱEastwoodȱJ,ȱetȱal.ȱEffectsȱofȱcaffeineȱ
onȱhumanȱhealth.ȱFoodȱAdditȱContamȱ2003;ȱ20(1):1Ȭ30.ȱ
[58] O’Brienȱ MC,ȱ McCoyȱ TP,ȱ Rhodesȱ SD,ȱ etȱ al.ȱ
Caffeinatedȱ cocktails:ȱ Energyȱ Drinkȱ Consumption,ȱ
HighȬRiskȱ
Drinking,ȱ
andȱ
AlcoholȬrelatedȱ
ConsequencesȱamongȱCollegeȱStudents.ȱAcadȱEmergȱ
Medȱ2008;ȱ15(5):453Ȭ460.ȱ
[59] Oddyȱ WHȱ andȱ O’Sullivanȱ TA.ȱ Energyȱ drinksȱ forȱ
childrenȱandȱadolescents.ȱBritȱMedȱJȱ2010;ȱ340(9):64.ȱ
[60] O’Deaȱ JA.ȱ Consumptionȱ ofȱ nutritionalȱ supplementsȱ
amongȱ adolescents:ȱ usageȱ andȱ perceivedȱ benefits.ȱ
HealthȱEducȱResȱ2003;ȱ18(1):98Ȭ107.ȱ
[61] Oteriȱ A,ȱ Salvoȱ F,ȱ Caputiȱ AP,ȱ etȱ al.ȱ Intakeȱ ofȱ Energyȱ
Drinksȱ inȱAssociationȱ withȱAlcoholicȱ Beveragesȱ inȱ aȱ
Cohortȱ ofȱ Studentsȱ ofȱ theȱ Schoolȱ ofȱ Medicineȱ ofȱ theȱ
Universityȱ ofȱ Messina.ȱ Alcoholȱ Clinȱ Expȱ Resȱ 2007;ȱ
31(10):1677Ȭ1680.ȱ
www.intechopen.com
!
[62] Programȱ onȱ Foodȱ Safety:ȱ Nutrientȱ additionȱ toȱ foodsȱ
inȱ Canada.ȱ Anȱ evaluationȱ ofȱ micronutrientȱ safety.ȱ
Departmentȱ ofȱ Nutritionalȱ Sciences,ȱ Universityȱ ofȱ
Torontoȱ1996.ȱ
[63] RagsdaleȱFR,ȱGronliȱTD,ȱBatoolȱN,ȱetȱal.ȱEffectȱofȱRedȱ
Bullȱ energyȱ drinkȱ onȱ cardiovascularȱ andȱ renalȱ
function.ȱAminoȱAcidsȱ2010;ȱ38(4):1193Ȭ1200.ȱ
[64] Rashtiȱ SL,ȱ Ratamessȱ NA,ȱ Kangȱ J,ȱ etȱ al.ȱThermogenicȱ
effectȱ ofȱ meltdownȱ RTDȱ energyȱ drinkȱ inȱ youngȱ
healthyȱ women:ȱ aȱ doubleȱ blind,ȱ crossȬoverȱ designȱ
study.ȱLipidsȱHealthȱDisȱ2009;ȱ8:57.ȱ
[65] Reissigȱ CJ,ȱ Strainȱ ECȱ andȱ Griffithsȱ RR.ȱ Caffeinatedȱ
energyȱ drinksȬAȱ growingȱ problem.ȱ Drugȱ Alcoholȱ
Depenȱ 2009;ȱ 99(1Ȭ3):1Ȭ10.Reynerȱ LAȱ andȱ Horneȱ JA.ȱ
Efficacyȱ ofȱ aȱ “functionalȱ energyȱ drink”ȱ inȱ
counteractingȱ driverȱ sleepiness.ȱ Physiolȱ Behavȱ 2002;ȱ
75(3):ȱ331Ȭ35.ȱ
[66] Riesenhuberȱ A,ȱ Boehmȱ M,ȱ Poschȱ M,ȱ etȱ al.ȱ Diureticȱ
potentialȱ ofȱ energyȱ drinks.ȱ Aminoȱ Acidsȱ 2006;ȱ
31(1):81Ȭ83.ȱ
[67] RiksenȱNP,ȱRongenȱGAȱandȱSmithsȱP.ȱPharmacologyȱ
andȱTherapeuticsȱ2009;ȱ121(2):ȱ185Ȭ191.ȱ
[68] SCFȱ(ScientificȱCommitteeȱonȱFoodȱAuthority),ȱ1999.ȱ
Opinionȱ onȱ caffeine,ȱ taurineȱ andȱ DȬglucuronoȬȱ
lactoneȱ asȱ constituentsȱ ofȱ soȬcalledȱ “energy”ȱ drinks,ȱ
adoptedȱonȱ21ȱJanuaryȱ1999.ȱ
[69] SCFȱ (Scientificȱ Committeeȱ onȱ Foodȱ Authority),ȱ 2003.ȱ
Opinionȱ ofȱ theȱ Scientificȱ Committeeȱ onȱ Foodȱ onȱ
Additionalȱinformationȱonȱ“energy”ȱdrinksȱ(expressedȱ
onȱ5ȱMarchȱ2003).ȱEuropeanȱCommission,ȱBrussels.ȱȱ
[70] Scholeyȱ ABȱ andȱ Kennedyȱ DO.ȱ Cognitiveȱ andȱ
physiologicalȱ effectsȱ onȱ anȱ “energyȱ drink”:ȱ anȱ
evaluationȱ ofȱ theȱ wholeȱ drinkȱ andȱ ofȱ glucose,ȱ
caffeineȱ
andȱ
herbalȱ
flavouringȱ
fractions.ȱ
Psychopharmacolȱ2004;ȱ176(3Ȭ4):320Ȭ330.ȱȱ
[71] Shaoȱ A.,ȱ andȱ Hathcockȱ JN.ȱ Riskȱ assessmentȱ forȱ theȱ
aminoȱ acidsȱ taurine,ȱ lȬglutamineȱ andȱ lȬarginine.ȱ Regȱ
ToxicolȱPharmacol,ȱ2008,ȱ50:ȱ376Ȭ399.ȱ
[72] Staibȱ AH,ȱ Stilleȱ W,ȱ Dietleinȱ G,ȱ etȱ al.ȱ Interactionȱ
betweenȱquinolonesȱandȱcaffeine.ȱDrugsȱ1987;ȱSupplȱ
1:170Ȭ174.ȱ
[73] StatisticsȱCanada,ȱTheȱCanadianȱCommunityȱHealthȱ
Survey,ȱ2004.ȱȱ
[74] Steinkeȱ L,ȱ Lanfearȱ DE,ȱ Dhanapalȱ V,ȱ etȱ al.ȱ Effectȱ ofȱ
“energyȱ drink”ȱ consumptionȱ onȱ hemodynamicȱ andȱ
electrocardiographicȱ parametersȱ inȱ healthyȱ youngȱ
adults.ȱAnnȱPharmacotherȱ2009;ȱ43(4):596Ȭ602.ȱ
ȱ
ȱ
ȱ
ȱ
www.intechopen.com
ȱ
[75] SvedȱDW,ȱGodseyȱJL,ȱLedyardȱSL,ȱetȱal.ȱAbsorption,ȱ
tissueȱ distribution,ȱ metabolismȱ andȱ eliminationȱ ofȱ
taurineȱ givenȱ orallyȱ toȱ rats.ȱ Aminoȱ Acidsȱ 2007;ȱ
32(4):4459Ȭ4466.ȱȱ
[76] TakahashiȱH,ȱKanedaȱS,ȱFukudaȱK,ȱetȱal.ȱStudiesȱonȱ
theȱteratologyȱandȱthreeȱgenerationȱreproductionȱofȱ
taurineȱinȱmice.ȱPharmocometricsȱ1972;ȱ6:535Ȭ540.ȱ
[77] TempleȱJL.ȱCaffeineȱuseȱinȱchildren:ȱWhatȱweȱknow,ȱ
whatȱ weȱ haveȱ leftȱ toȱ learn,ȱ andȱ whyȱ weȱ shouldȱ
worry.ȱNeurosciȱBiobehavȱRȱ2009;ȱ33(6):793Ȭ806.ȱ
Theȱ Expertȱ Groupȱ onȱ Vitaminsȱ andȱ Minerals:ȱ Safeȱ
UpperȱLevelsȱforȱVitaminsȱandȱMinerals.ȱLondonȱ(UK):ȱ
Foodȱ Standardsȱ Agency,ȱ Expertȱ Groupȱ onȱ Vitaminsȱ
andȱ Mineralsȱ Mayȱ 2003.ȱ http://www.food.gov.uk/ȱ
multimedia/pdfs/vitmin2003.pdfȱ
[78] ThombsȱDL,ȱO’MaraȱRJ,ȱTsukamotoȱM,ȱetȱal.ȱEventȬ
levelȱ analysesȱ ofȱ energyȱ drinkȱ consumptionȱ andȱ
alcoholȱ intoxicationȱ inȱ barȱ patrons.ȱ Addictȱ Behavȱ
2010;ȱ35(4):325Ȭ330.ȱ
[79] Timbrellȱ JA,ȱ Seabraȱ V,ȱ andȱ Waterfieldȱ CJ.ȱ Theȱ inȱ
vivoȱanȱdinȱvitroȱprotectiveȱproertiesȱofȱtaurine.ȱGenȱ
Pharmacolȱ1995,ȱ26,ȱ453Ȭ462.ȱȱ
[80] USDAȱ2011.ȱhttp://ndb.nal.usda.gov/ndb/search/list.ȱ
[81] Warburtonȱ DM,ȱ Berselliniȱ Eȱ andȱ Sweeneyȱ E.ȱ Anȱ
evaluationȱ ofȱ aȱ caffeinatedȱ taurineȱ drinkȱ onȱ mood,ȱ
memoryȱ andȱ informationȱ processingȱ inȱ healthyȱ
volunteersȱ
withoutȱ
caffeineȱ
abstinence.ȱ
Psychopharmacolȱ2001;ȱ158(3):322Ȭ328.ȱ
[82] Warskulatȱ U.,ȱ HellerȬStilbȱ B.,ȱ Oermannȱ E.ȱ Etȱ al.,ȱ
Phenotypeȱ ofȱ theȱ taurineȱ transporterȱ knockoutȱ
mouse.ȱMethodsȱEnzymolȱ2007;ȱ428;ȱ439Ȭ458.ȱ
[83] Wooȱ J.,ȱ 2007.ȱ Adverseȱ eventȱ monitoringȱ andȱ
multivitaminȬmultimineralȱ dietaryȱ supplements.ȱ
Am.ȱJ.ȱClinicalȱNutrition.ȱ85(1):323SȬ324S.ȱ
[84] Worthleyȱ MI,ȱ Prabhuȱ A,ȱ Deȱ Sciscioȱ P,ȱ etȱ al.ȱ
Detrimentalȱeffectsȱofȱenergyȱdrinkȱconsumptionȱonȱ
plateletȱ andȱ endothelialȱ function.ȱ Amȱ Jȱ Medȱ 2010;ȱ
123(2):184Ȭ187.ȱ
[88]ȱ Zucconi,ȱ S.,ȱ Volpato,ȱ C.,ȱ Adinolfi,ȱ F.,ȱ etȱ al.,ȱ
Gatheringȱ consumptionȱ dataȱ onȱ specificȱ consumerȱ
groupsȱ ofȱ energyȱ drinks.ȱ Anȱ externalȱ scientificȱ
report.ȱEuropeanȱFoodȱSafetyȱAuthroity,ȱ2013.ȱ
ȱ
ȱ
ȱ
29

Documents Submitted to the Committee (Part 2)

Transcription

Similar documents

March 2016 - River Bend Retirement Community

Exam Stress Brochure - AB Health Services

MHM 2016 Poster

The Energy Drink Buzz

Hello Club Members! - Seventh Generation Club

Teen Fads 2

North West Vending Bolero

James R. Ginder, MS,NREMT,PI,CHES,NCEE

Caffeine-Coffee- An Essay by Malcolm Gladwell