Systematic review of the potential adverse effects
of caffeine consumption in healthy adults, pregnant
women, adolescents, and children:
Cardiovascular Outcome
DR. JEFFREY GOLDBERGER
UNIVERSITY OF MIAMI MILLER SCHOOL OF MEDICINE
PRESENTED BY
D R . D A N I E L E W I KO F F
T O X S T R AT E G I E S
1
PECO Questions for the Cardiovascular Outcome
For [population], is caffeine intake above [dose], compared to intakes [dose]
or less, associated with adverse effects on cardiovascular health outcomes?
Population
Healthy
Adults
Healthy
Pregnant Women
Healthy
Adolescents
Healthy
Children
Exposure
> 400 mg/day
> 300 mg/day
> 2.5 mg/kg-day
> 2.5 mg/kg-day
Comparator
≤ 400 mg/day
≤ 300 mg/day
≤ 2.5 mg/kg-day
≤ 2.5 mg/kg-day
Outcome
Adverse cardiovascular effects
Example: For healthy adults, is caffeine intake above 400 mg/day, compared to
intakes 400 mg/day or less, associated with adverse effects on cardiovascular
health outcomes?
2
Characterization of Data for Cardiovascular Outcome
Data Type
Characterization
Number of studies
203 included; 73 excluded
Study types
Mostly randomized controlled trials; some observational studies; 5 meta-analyses
Populations
Mostly healthy adults; few children/adolescents
Exposure
Pure caffeine (RCTs) or
Dietary, e.g., coffee, tea soda, energy drinks etc. (observational studies)
Outcome (Endpoints)
Clinical: Mortality (e.g. stroke, coronary heart disease)
Morbidity (e.g. acute myocardial infarctions, atrial fibrillation, stroke)
Physiological: aortic stiffness, blood pressure, cerebral blood flow, heart rate, heart
rate variability, LDL, cholesterol, plasma and urinary constituents, other
hemodynamic measurements, ventricular function
3
Mortality
The majority of evidence support that 400 mg caffeine/day in healthy adult populations is an acceptable intake
which is not associated with significant concern for cardiovascular mortality
◦ At higher intakes up to ~855 mg/day, there are no consistently reported effects
◦ Several studies, reported findings that are suggestive of protective effects
◦ Studies reporting effects, both above and below the comparator, were conditional (e.g., observed in subset of data evaluated)
Moderate level of confidence in the body of evidence
4
Morbidity
Evidence support that 400 mg caffeine/day in healthy adult populations is an acceptable intake which is not
associated with significant concern regarding cardiovascular morbidity
◦ Several studies (including meta-analyses) suggested that the comparator is too low
◦ All comparison points above 400 mg/day (up to 1050 mg/day) were no effect levels
◦ Some studies, including two meta-analyses reported effect levels below the comparator
◦ Associations were observed only in specific genotypes (highlighting potential role of kinetic influence)
Moderate level of confidence in the evidence base
◦ increased by the low level of indirectness, low risk of bias in the individual studies
5
Blood Pressure
Intakes both below and above the comparator, have the
potential to result in an increase in blood pressure (often
only a few mmHg) in all populations evaluated
Biological significance of this small magnitude of change
is difficult to interpret (and is conditional)
◦ The range of normal blood pressure variation during the day
exceeds the increase that is associated with caffeine
◦ The blood pressure increase with exercise is typically
substantially greater than that observed with caffeine
The comparator of 400 mg/day in healthy adults is too high if one is only considering the potential for caffeine
to cause a physiological change in blood pressure
The comparator of 400 mg/day is likely acceptable
◦ Considering the small magnitude of changes in this physiological parameter
◦ Lack of information demonstrating an association between chronic caffeine-mediated blood pressure increases relative
to known cardiovascular risk factors
Moderate level of confidence in this evidence base
6
Evaluation of Individual
Study Quality (Risk of
Bias)
Overall “probably low”
risk of bias
Limited by exposure
characterization in
observational studies (Q8)
+2: Definitely low
+1: Probably low
-1: Probably high
-2: Definitely high
7
Factors that increased or decreased confidence in the body of evidence
by endpoint
Endpoint
No. of
Studies
Initial
Confidence
Rating
Overall RoB
Consistency
Final Confidence
Rating
Magnitude
Confounding
Were plausible
confounders
that would
change the
observed effect
accounted for?
Were findings
consistent in
demonstrating
effects or lack of
effects at or
below the
comparator
What is the overall
rating when factors
that increase or
decrease
confidence were
considered
Indirectness
Based on study
type and study
features (OHAT,
2015)
Domainbased
evaluation of
risk of bias
per the OHAT
RoB tool
(OHAT, 2015)
Was the study
designed to
evaluate the
PECO?
Strength of
effect (when
effect
observed
below the
comparator)
Adults
Mortality
13
Moderate
↑/-
↑
--
↑
↑/-
Moderate
Morbidity
18
Moderate
↑/-
↑
--
↑
↓
Moderate
Blood pressure
115
Moderate to high
↑/-
↑
↓/-
↑
--
Moderate
Heart rate
53
Moderate to high
↑/-
↑
Cholesterol
24
Moderate to high
↑/-
↑/-
↓
—
↑
Moderate to high
Heart rate viability
13
Moderate to high
↑/-
↑
--
—
↑/-
Moderate to high
Blood pressure
10
Moderate to high
↑/-
↑
↓
—
↑
High
Heart rate
6 increased confidence
High
↑
—
Adolescents and children
-↑no change to ↑
confidence
—
— confidence ↑/↓ decreased
High
8
Developing Conclusions for the Outcome: Integration of Weight of
Evidence Considerations
9
WoE Conclusion: Cardiovascular
Adults: When the weight of evidence was considered, with particular emphasis on level of
adversity, 400 mg caffeine/day was found to be an acceptable intake which is not associated
with significant concern regarding adverse cardiovascular effects.
◦ Some findings suggested that the comparator was too low (clinical endpoints)
◦ Some findings reported effects below the comparator (mostly physiological endpoints)
◦ Limited by the inability to ascertain the conditions and magnitude of change that would be considered adverse Moderate to high
level of confidence in the evidence base.
Children and adolescents: the evidence base was insufficient to develop conclusion
◦ Available data (blood pressure and heart rate) are inconsistent
10
Future Research Needs
More research in children and adolescent populations
Better understanding of dose-response relationships following chronic exposure for some
endpoints (e.g. endothelial function and heart rate variability)
In some cases (e.g. certain physiological endpoints), a better characterization of what, if any,
magnitude of change may be associated with adverse effects
11