Controversies in Nutrition
By Alan R. Gaby, M.D.
A Modern Day Tea Party?
Don’t be Mad: Be Responsible
Recent controversies
Does calcium cause heart disease?
Is supplementing with high-dose vitamin D to
achieve an “optimal” serum 25(OH)D level safe
and beneficial?
Is megadose iodine safe and beneficial?
Does vitamin E increase mortality and cause
congestive heart failure?
Recent controversies
Is lowering homocysteine levels with B vitamins
beneficial?
Does folic acid cause cancer?
Does vitamin A cause osteoporosis and increase
mortality?
Is high-dose strontium safe and effective for
osteoporosis?
Does calcium cause heart disease?
Meta-analysis of 15 randomized controlled
trials: participants who received
supplemental calcium as monotherapy (i.e.,
without other nutrients) had a 30% increase
in the incidence of myocardial infarction
(p = 0.035 to 0.038).
BMJ 2010;341:c3691
Does calcium cause heart disease?
The data were derived from secondary
(post hoc) analyses of studies (mainly
osteoporosis studies) that were not designed
to test the effect of calcium on heart disease
risk.
BMJ 2010;341:c3691
Does calcium cause heart disease?
Findings of borderline statistical significance
from post hoc analyses are more likely to be
due to chance than are findings of borderline
statistical significance from primary
analyses.
Calcium-magnesium interrelationships
(Effects of a high-calcium diet in rats and pigs)
Decreased tissue magnesium levels
Fed Proc 1986;45:374
Increased magnesium requirements
J Nutr 1960;70:103-111
Increased severity of magnesium deficiency
in animals fed a magnesium-deficient diet
Am J Physiol 1951;166:408-12
Calcium-magnesium interrelationships
(Effects of high-calcium intake in humans)
2 g/day of calcium (citrate) decreased Mg
absorption and plasma Mg levels in healthy
volunteers. Clin Sci 1967;32:11-18
Calcium supplementation had no effect on
Mg balance in adolescent girls.
Am J Clin Nutr 1996;63:950-3
Magnesium: a cardioprotective nutrient
Inhibits platelet aggregation
Vasodilator
Anti-arrhythmic activity
Required for ATP synthesis
Promotes intracellular potassium uptake
Possibly lowers BP and increases HDL-C
Magnesium: a cardioprotective nutrient
Rats fed a Mg-deficient diet developed myocardial
necroses. Am J Pathol 1964;45:757-68
In rats, epinephrine-induced myocardial necroses were
prevented by Mg. Arzneimittelforschung 1983;33:205-10
Mg prevented myocardial infarction induced by coronary
artery ligation in rats. Can Med Assoc J 1960;82:212-3
Mg prevented the development of atherosclerosis in
animals fed an atherogenic diet. Proc Natl Acad Sci 1990;87:1840-4
Magnesium intake is frequently low
NHANES 1999-2000: 50% of Caucasians
consumed < 75-80% of the RDA; Mg intake was
about 25% lower in African-Americans than in
Caucasians.
J Nutr 2003;133:2879-82
Mean Mg intake by high school and college
women was 125 mg/day (60-65% below the
RDA)
J Am Diet Assoc 1969;55:38-43
Calcium-magnesium interrelationships
In people with low or suboptimal Mg status,
administration of calcium without
concomitant Mg supplementation could
further compromise Mg status, and thereby
increase the risk of developing heart disease.
Calcium-silicon interrelationships
In rats, calcium supplementation decreased
the silicon content of bone. Fed Proc 1970;29:565
As a cross-linking agent, silicon may help
protect arteries from injury. Lancet 1977;1:454-7
Silicon prevented the development of
atherosclerosis in rabbits fed an atherogenic
diet. Atherosclerosis 1979;33:397-408
Calcium: to supplement or not?
Adequate calcium intake is essential for
optimal bone health.
In most instances, calcium supplementation
should be accompanied by Mg (usual ratio,
approximately 2:1), and possibly by silicon
(perhaps 2-5 mg/day, as part of a multimineral formula).
High-dose vitamin D:
Is it safe
and effective?
Vitamin D deficiency
Rickets
Osteomalacia, osteoporosis
Weakness, falls
Myalgia
Decreased resistance against infection
Benefits of supplementation
Fewer falls & fractures, better bone density
Greater insulin sensitivity
Prevention of influenza
Prevention of asthma attacks
Improvement of hypertension?
Better mood?
Vitamin D: effective dosages
800-1,200 IU/day generally effective
400 IU/day generally ineffective
New RDA (2010): 600 IU/day for ages 1-70;
800 IU/day for ages ≥ 71
2,000 IU/day was used for prostate cancer:
slowed disease progression, decreased pain.
Purported benefits of maintaining
“optimal” vitamin D status
“Optimal” outcomes re falls, fractures,
insulin sensitivity, BP, mood, etc.
Prevention of some cancers
Prevention of some autoimmune diseases
Better dental health
Vitamin D: new definition of
deficiency
Traditional definition:
deficiency = serum 25(OH) < 10-15 ng/ml
(< 25-37.5 nmol/L)
New definition:
deficiency = serum 25(OH)D < 20 ng/ml
(< 50 nmol/L)
insufficiency = < 30 ng/ml (< 75 nmol/L)
Vitamin D: new definition of “optimal”
A review article concluded that a protective
effect with respect to various outcomes
began at a serum 25(OH)D level of 30 ng/ml
(75 nmol/L) and that the best outcomes were
seen in people with levels of 36-40 ng/ml
(90-100 nmol/L).
Am J Clin Nutr 2006;84:18-28
Dosage requirements for new
“adequate” and “optimal”
Only 50% of people will achieve “adequacy” (≥ 30 ng/ml)
with 1,000 IU/day.
1,644-3,440 IU/day (depending on the study) will achieve
“adequacy” in nearly all healthy adults.
Even larger doses (4,000-10,000 IU/day?) needed to
achieve “optimal” levels.
Tolerable Upper Intake Level = 4,000 IU/day (recently
increased from 2,000 IU/day)
Examining the evidence
Is routine use of vitamin D in dosages
greater than 2,000 IU per day
beneficial?
Is it safe?
My conclusions
Serum 25(OH)D is not always a reliable
indicator of vitamin D status.
The new definitions of vitamin D deficiency and
insufficiency may not be valid for some
populations and some individuals.
Evidence supporting the benefit of pushing
25(OH)D to an “optimal” level is weak.
My conclusions
Evidence supporting the long-term safety
of dosages > 2,000/day is weak.
The safety and efficacy of vitamin D
supplementation cannot be inferred from
data regarding the safety and efficacy of
sunlight exposure.
Assessment of vitamin D status
Serum vitamin D: unreliable; serum half-life is
only 24 hours.
Serum 1,25-dihydroxyvitamin D: unreliable;
may be normal or elevated in people with
vitamin D deficiency. Am J Clin Nutr 2004;79:362-371
Serum 25-hydroxyvitamin D: serum half-life
is 3 weeks; more reliable than vitamin D
itself.
Assessment of vitamin D status
Serum 25-hydroxyvitamin D
↓?
Serum 25-hydroxyvitamin D
Accuracy of 25(OH)D measurement
Substantial variations from one lab to
another and with different assay methods
With nearly identical serum samples, one
lab found that 90% were below 32 ng/ml;
another lab found that only 17% were
below 32 ng/ml.
Am J Clin Nutr 2008;87:1087S-91S
Validity of 25(OH)D measurement
25(OH)D is only one of more than 50 vitamin D
metabolites identified.
Vitamin D nutritional status may be a function
of complex interactions between many different
vitamin D metabolites.
Different people may have different serum
25(OH)D “set points” for adequate or “optimal”
vitamin D nutritional status.
Validity of 25(OH)D measurement
Serum 25(OH)D may be even less reliable as an
indicator of vitamin D status when vitamin D
doses are greater than 2,000 IU/day, because
25-hydroxylases become saturated at those
dosages. Storage of large amounts of
unmetabolized vitamin D may not be reflected
in serum 25(OH)D measurements.
Am J Clin Nutr 2008;87:1738-42
New definition of deficiency: is it valid?
Definition is based on biochemical
markers:
As 25(OH)D levels go up, parathyroid
levels tend to go down and fractional
calcium absorption tends to increase.
New definition of deficiency: is it valid?
Vitamin D sufficiency is inferred when a further
increase in serum 25(OH) does not further
depress parathyroid hormone levels or further
increase fractional calcium absorption. In
population studies, the average 25(OH)D level
at which vitamin D sufficiency occurred was
around 30 ng/ml (75 nmol/L).
N Engl J Med 2007;357:266-81
New definition of deficiency: is it valid?
In the absence of severe vitamin D
deficiency, the associations between serum
25(OH), parathyroid hormone, and
fractional calcium absorption do not seem
to apply to all population groups, nor to all
individuals.
New definition of deficiency: is it valid?
Of 93 young adults living in Hawaii who
had sun exposure a mean of 29 hours a
week, 25-51% had a 25(OH)D level < 30
ng/ml and 3-8% had a level < 20 ng/ml.
There was no correlation between
25(OH)D and parathyroid hormone levels.
J Clin Endocrinol Metab 207;92:2130-5
New definition of deficiency: is it valid?
Those findings suggest either that the
cut-off level for 25(OH)D used to define
vitamin D sufficiency is either
inappropriately high for some groups or
that 25(OH)D is not always a reliable
indicator of vitamin D nutritional status.
J Clin Endocrinol Metab 207;92:2130-5
New definition of deficiency: is it valid?
In the late 1990s, the standard RIA for
25(OH)D was changed. The new method
decreased measured values by 4 ng/ml
(10 nmol/L). Am J Clin Nutr 2008;88:1519-27
The new cut-off points for deficiency and
insufficiency were based in part on studies
done prior to the late 1990s.
New definition of “optimal”: is it valid?
Evidence is derived mainly from observational
studies in which serum 25(OH)D was correlated
with health outcomes. Findings conflicting.
Evidence is also derived from controlled trials in
which vitamin D-supplemented patients who
achieved higher 25(OH) levels had better
outcomes than did supplemented patients whose
25(OH)D levels were lower.
Limitations of observational studies
Failure to control for confounders such as age,
BMI, co-morbidities
High 25(OH)D levels result mainly from
sunlight exposure. People who spend time in the
sun differ from those who do not.
If sun exposure is beneficial, the effect may not
be due entirely (or even primarily) to vitamin D.
Findings from controlled trials
Clinical trials did not compare higher-dose
versus lower-dose vitamin D.
Instead they assessed outcomes as a function of
the serum 25(OH)D response to vitamin D
supplementation. That measurement might
simply be identifying differences in body
chemistry, rather than an effect of vitamin D
supplementation.
Findings from controlled trials
A higher serum 25(OH)D response to
supplementation might reflect:
More efficient nutrient absorption in
general
More efficient 25-hydroxylation of
vitamin D
Hepatic hydroxylase enzymes
Four different cytochrome P450 enzymes are
thought to be capable of 25-hydroxylating
vitamin D. Trends Biochem Sci 2004;29:664-73
Cytochrome P450 enzymes also help detoxify
xenobiotic chemicals.
Hydroxylase enzymes also play a role in the
synthesis of DHEA and estriol, both of which
may have health benefits.
Extra-hepatic hydroxylase enzymes
Human testis (androgen-producing Leydig cells)
and possibly ovary are also capable of
25-hydroxylating vitamin D.
25(OH)D levels were 60% lower in young men
with h/o orchiectomy for bilateral testicular
cancer than in matched controls.
Lancet 2010;376:1301
Extra-hepatic hydroxylase enzymes
Observational studies on 25(OH)D levels
and health outcomes may be confounded
by differences in levels of testosterone and
DHEA. Both of these androgenic
hormones may have positive influences on
health.
Is high-dose vitamin D safe?
Tolerable Upper Intake Level for adults is
4,000 IU per day (recently increased from
2,000 IU/day).
Some investigators have argued that up to
10,000 IU per day is safe for most adults.
Basis of the argument that long-term use
of 10,000 IU/day of vitamin D safe
Hypercalcemia uncommon with 10,000
IU/day
Whole-body sun exposure results in
production of at least 10,000 IU/day
without causing vitamin D toxicity.
Weaknesses of the argument
1. High-dose supplementation studies were of
short duration.
2. Absence of hypercalcemia is not proof of safety.
3. Unclear whether human skin really can produce
10,000 IU/day of vitamin D
4. Physiological effects of sunlight exposure differ
from those of vitamin D supplementation.
High-dose supplementation studies
were of short duration
10,000 IU/day was given for a maximum of
20 weeks. As a fat-soluble nutrient, vitamin
D can accumulate with continued
administration.
Absence of hypercalcemia is not proof
of safety
An increase in urinary calcium excretion (even
within the normal range) might increase the risk of
developing kidney stones.
3 of 45 elderly individuals who received 5,000
IU/day of vitamin D for 12 months showed
evidence of hypercalciuria. Am J Clin Nutr 2009;89:1132-7
Absence of hypercalcemia is not proof
of safety
Swine fed human equivalent of 11,500 IU/day of
vitamin D3 developed pathological changes in the
aorta that were indistinguishable from human
atherosclerosis. Am J Clin Nutr 1979;32:58-83
Increasing vitamin D3 intake only modestly
(equivalent to a total of 917 IU/day for humans)
exacerbated atherosclerosis in swine induced by a
diet high in butterfat. Nutr Rep Int 1983;28:1111-8
Can human skin can produce 10,000
IU/day?
This claim is based in part on a study in which UV
irradiation of 5% of body surface area was
equivalent to oral administration of 400 IU/day.
J Bone Miner Res 1998;13:1238-42
No evidence that it is appropriate to extrapolate
this finding to full-body irradiation
Can human skin can produce 10,000
IU/day?
One-time exposure to 1 minimal erythemal dose of
UV irradiation was equivalent to oral
administration of 10,000-25,000 IU of vitamin D2.
This finding is of doubtful relevance to long-term
vitamin D homeostasis.
Can human skin can produce 10,000
IU/day?
Repeated sun exposure results in photodegradation
of vitamin D that has not yet entered the
circulation. Am J Clin Nutr 1995;61(Suppl):638S-45S
Therefore, net vitamin D production may be
substantially lower on subsequent days than on the
first day.
UV light and oral vitamin D are not
the same
One photodegradation product of vitamin D
(5,6-trans-vitamin D) has effects similar to 1,25dihdroxyvitamin D, but is 20-40 times less potent.
Biochemistry 1972;11:2715-9
5,6-trans-Vitamin D might compete with
1,25(OH)2D and thereby function as a regulator of
vitamin D activity.
UV light and oral vitamin D are not
the same
Sunlight (but not vitamin D):
Produces photodegradation products
Produces corticotropin-releasing hormone
May directly influence hypothalamic and
pituitary function through the retina
What to make of it all
RDAs of 400-600 IU/day are not sufficient to
promote optimal health. 800-1,200 IU/day is more
effective than 400 IU/day.
The safety and efficacy of using high doses (such
as > 2,000 IU/day) for the sole purpose of
achieving a target 25(OH)D level has not been
established.
What to make of it all
Doses greater than 2,000 IU per day may be
considered in selected cases, but the potential
benefits should be weighed against the risks.
Sunlight 5-15 min 2-3 times a week between
10 a.m. and 3 p.m. in spring, summer, and
autumn is usually sufficient for skin types II
and III. Am J Clin Nutr 2004;80(Suppl):1678S-88S
Iodine facts (µg/day)
Adult RDA
150
Median urinary [I] in US adults
168
Tolerable Upper Intake Level
1,100
Iodine: adverse effects
Very high doses 700-4,500 mg/day)
Thyroid dysfunction (mainly hypothyroidism), burning mouth, increased
salivation, parotid and submandubular
swelling, severe headache, acneiform
eruptions, pulmonary edema, angioedema,
heart failure, and death.
Iodine: adverse effects
Moderately high doses (3-6 mg/day)
10.9% of 1,365 women treated for
fibrocystic breast changes experienced side
effects including acne, nausea, diarrhea,
thinning hair, skin rash, headache,
hypothyroidism (0.3%), and hyperthyroidism
(0.1%). Can J Surg 1993;36:453-60
Iodine: adverse effects
Modestly high doses (> 500 µg/day?)
Autoimmune thyroiditis
Hypothyroidism
Goiter or increased thyroid volume
N Engl J Med 2006;354:2783-93; Thyroid 2003;13:561-7; Clin Endocrinol
1991;34:413-6; Lancet 1987;2:257-9; Am J Clin Nutr 2005;81:840-4
Iodine:
A Lot
to Swallow
“Orthoiodosupplementation”
(Guy Abraham, M.D.)
According to Abraham, the optimal dietary
iodine intake is 13.8 mg/day, which is 92
times the RDA and more than 12 times the
Tolerable Upper Intake Level.
http://www.optimox.com/pics/Iodine/IOD-02/IOD_02.htm
Basis of the claim
Japanese people consume an average of 13.8
mg/day of iodine, and are among the
healthiest people in the world.
High-doses are needed to fully saturate the
tissues, as demonstrated by an iodine-load
test.
But it doesn’t make sense!
Of the many criticisms I had of Abraham’s
theory, one was related to evolution: We have
evolved in an iodine-poor environment. Our
thyroid glands have developed a powerful
mechanism to concentrate iodine, and some
thyroid glands (or other tissues) might not
function properly if iodine intake is increased
by 90-fold.
To which Abraham replied . . .
“It is understandable why someone who believes
in the theory of evolution has a problem with
such high requirements for iodine in an
environment depleted of this element. According
to the biblical narrative, the Creator declared
planet earth and everything in it perfect.
Therefore, the original planet earth contained
a topsoil rich in iodine, and all elements
required for perfect health of Adam, Eve, and
their descendants.”
To which Abraham replied . . .
“A sequence of events followed, culminating in
the worldwide flood 4,500 years ago. Following
this episode, the receding waters washed away
the topsoil with all its elements into oceans and
seas. The new topsoil became deficient in iodine
. . .”
Townsend Letter for Doctors & Patients, October 2005
That doesn’t make sense either!
Humans have a powerful mechanism to
concentrate iodine in the thyroid gland. In
an iodine-rich antediluvian world, that
mechanism would have been an
imperfection, so the Creator could not have
declared everything perfect.
Other statements by Abraham . . .
“Following World War II, there was a systematic
attempt to remove iodine from the food supply of
Christian America. Iodophobic misinformation,
well synchronized with the introduction of
alternatives to iodine supplementation in medical
practice, strongly suggest a well planned
conspiracy by agents of foreign powers planted at
strategic positions in academia and the regulatory
agencies.”
http://www.optimox.com/pics/Iodine/pdfs/IOD16.pdf. Accessed 10/10/10
Other statements by Abraham . . .
“In the early 1960’s, potassium iodate was added
to bread as a dough conditioner . . . . This was an
oversight by the agents of foreign powers . . . .
One slice of bread contained the full RDA of
150 mcg. This amount of the dezombifier iodine
in a major staple food of Christian America could
not be tolerated for long.”
http://www.optimox.com/pics/Iodine/pdfs/IOD16.pdf. Accessed 10/10/10
Foreign agents plotting deiodination
of America
Do Japanese people consume 13.8 mg/day?
Claim based on a misinterpretation of a 1967
paper. Average seaweed consumption in Japan =
4.6 g/day. Seaweed contains average of 0.3%
iodine.
4,600 mg x 0.003 = 13.8 mg
However, 4.6 g/day of seaweed was wet weight,
whereas 0.3% iodine was based on dry weight.
J Clin Endocrinol Metab 1967;27:638-47
Amount of iodine consumed in Japan
In studies in the 1990s that specifically
looked at iodine intake in Japan, mean
dietary iodine (estimated from urinary
iodine excretion) was 330-500 µg/day,
which is 25-fold lower than 13.8 mg/day.
Nippon Naibunpi Gakkai Zasshi 1994;70:1093-1100; Nippon Naibunpi Gakkai
Zasshi 1992;68:550-6
Amount of iodine consumed in Japan
According to a 2008 study, average iodine
intake in Japan from seaweed was 1.2
mg/day in 2006 and 1.7 mg/day in 1986,
which is 88-93% less than 13.8 mg/day.
Thyroid 2008;18:667
Abraham’s iodine load test
Patient ingests 50 mg of iodine/iodide.
Patient considered iodine-deficient if < 90%
is excreted in the urine over the next 24
hours.
92-98% of patients taking the test have been
found to be deficient.
Abraham’s iodine load test
The validity of the test depends on the
assumption that the average person can
absorb at least 90% of a 50-mg dose.
No research in humans; proponents have
not measured fecal iodine levels. In cows
fed supraphysiological doses of iodine, 50%
appeared in the feces. J Dairy Sci 1996;79:254-9
Adverse effects of
“orthoiodosupplementation”
Toxic multinodular goiter
Graves’ disease
Autoimmune thyroiditis
Hypothyroidism
Severe headaches
Deep acne
Hair loss, agitation, sweating
Esophagitis
Megadose iodine: conclusion
3-6 mg/day may be considered for fibrocystic
breast changes that do not improve with
methylxanthine avoidance, vitamin E, etc.
As an antimicrobial agent, iodine may produce
clinical benefit in selected patients by killing
intestinal pathogens.
Beneficial for some other conditions, such as
erythema nodosum and possibly some types of
cysts.
Megadose iodine: conclusion
Iodine is not indicated as a treatment for
hypothyroidism except in cases of dietary
iodine deficiency. High iodine intake can
make hypothyroidism worse.
Megadose iodine: conclusion
There is no credible evidence that routinely
giving high-dose iodine based on an iodine
load test or on a misunderstanding of human
iodine requirements is either safe or
beneficial.
Side effects of high-dose iodine are common,
and in a small proportion of cases side effects
are severe and/or persistent.
Does vitamin E cause heart failure?
In a double-blind study of patients with
vascular disease or diabetes, treatment with
400 IU/day of alpha-tocopherol for 7 years
resulted in a significant 19% increase in the
incidence of heart failure compared with
placebo.
JAMA 2005;293:1338-47
Does vitamin E increase mortality?
Meta-analysis of 19 randomized controlled
trials found that long-term supplementation
with ≥ 400 IU/day of vitamin E was associated
with a significant 4% increase in all-cause
mortality compared with placebo. In contrast,
< 400 IU/day was associated with a
nonsignificant decrease in all-cause mortality.
Ann Intern Med 2005;142;37-46
Vitamin E and heart failure:
possible explanation
The study used pure alpha-tocopherol,
whereas vitamin E as it occurs naturally in
food consists of 4 isomers: alpha-, beta-,
gamma-, and delta-tocopherol. Supplementation with large doses of alpha-tocopherol
(AT) can deplete gamma-tocopherol (GT).
J Nutr 2003;133:3137-40
Importance of gamma-tocopherol (GT)
GT is metabolized largely to 2,7,8-trimethyl-2-(beta-carboxyethyl)-6-hydroxychroman (gamma-CEHC), which may
function as a natriuretic hormone, since it is
involved in the body’s response to sodiuminduced plasma volume expansion.
J Pharmacol Exp Ther 1997;282:657-62; Proc Natl Acad Sci 1996;93:6002-7;
J Nutr Sci Vitaminol 2004;50:277-82
Importance of gamma-tocopherol (GT)
GT depletion might lead to impaired regulation
of sodium and water balance, potentially
increasing the risk for heart failure.
If high-dose AT does adversely affect heart
function in some people, one might reasonably
expect that mixed tocopherols (which contain AT
and GT) would not have the same negative effect,
and might even be beneficial.
Vitamin E and mortality: study flaws
Three of the 19 studies in the meta-analysis,
which included a total of 27,000
participants (two-thirds of all of the patients
in the pooled analysis) had major
weaknesses.
Vitamin E and mortality: study flaws
First study (n = 20,536): Patients, most of
whom had heart disease, received betacarotene in addition to vitamin E.
Presumably, a large proportion were
smokers. Beta-carotene increases incidence
of lung cancer in smokers.
Lancet 2002;360:23-33; N Engl J Med 1994;330:1029-35
Vitamin E and mortality: study flaws
Second study (n = 3,640): In addition to receiving
vitamin E, patients received 80 mg/day of zinc
for an average of 6.3 years.
Arch Ophthalmol 2001;119:1417-36
Long-term use of high-dose zinc can cause
copper deficiency, which can lead to heart
disease. Copper was given, but as cupric oxide,
which is not absorbable.
Vitamin E and mortality: study flaws
Third study (n = 2,002): Vitamin E and
placebo groups were not comparable at
baseline. Vitamin E group had significantly
higher serum cholesterol and significantly
higher prevalence of hypertension, diabetes,
smoking, and severe coronary artery
disease.
Lancet 1996;347:781-6
If vitamin E does increase mortality
Possible explanations:
Use of synthetic (all-rac, or DL), as
opposed to “natural” (RRR or D), alphatocopherol.
Use of alpha-tocopherol instead of mixed
tocopherols
Vitamin E: synthetic vs. natural
Synthetic (all-rac, or DL): Only 1 of the 8
isomers is found naturally in the body.
In the only high-dose study in the metaanalysis (other than the flawed studies) that
used “natural” (RRR, or D) alpha-tocopherol,
the mortality rates in the vitamin E and
placebo groups were identical.
N Engl J Med 2000;342:154-160
Vitamin E: alpha vs. mixed tocopherols
High-dose AT may deplete GT.
In addition to functioning as a natriuretic
hormone, GT was more effective than AT in
inhibiting platelet aggregation, prostate cancer
cell growth, and oxidative DNA damage, and in
increasing superoxide dismutase activity.
J Nutr 2003;133:3137-40; J Am Coll Cardiol 1999;34:1208-15
Vitamin E: alpha vs. mixed tocopherols
GT was more effective than AT as a scavenger
of peroxynitrite (a mutagenic nitrating and
oxidizing agent formed during the activation of
phagocytes. Peroxynitrite is believed to play a
role in the pathogenesis of cardiovascular
disease, cancer, and neurodegenerative
diseases.
Proc Natl Acad Sci 1997;94:3217-22
Vitamin E and mortality
If high-dose alpha-tocopherol (synthetic or
natural) does increase mortality slightly, one
might reasonably expect that the same
adverse effect would not occur with mixed
tocopherols, which contains both alpha- and
gamma-tocopherol.
Is lowering homocysteine beneficial?
Homocysteine is a byproduct of methionine
metabolism.
Oral or parenteral administration of
homocysteine thiolactone (a precursor to
homocysteine) caused atherosclerosis in
rabbits.
Atherosclerosis 1975;22:215-227
Homocysteine: adverse effects
Individuals with homozygous
homocystinuria (a rare inborn error of
metabolism) develop early-onset
osteoporosis, premature atherosclerosis,
and thromboembolism, and usually die
by age 13.
Homocysteine: observational studies
Hyperhomocysteinemia has been
associated with an increased incidence
of atherosclerosis, thromboembolism,
stroke, osteoporosis, recurrent
miscarriage, and age-related cognitive
decline.
Homocysteine: genetic factors
The 677C→T variant of methylenetetrahydrofolate reductase (which is associated
with hyperhomocysteinemia) is associated
with increased risk of heart disease, stroke,
dementia, miscarriage, osteoporosis. This
suggests that the association of hyperhomocysteinemia with these diseases is
causal.
Homocysteine: positive clinical trials
In hyperhomocysteinemic stroke patients,
5 mg/day of folic acid plus 1,500 µg/day of
vitamin B12 for 2 years reduced hip fracture
incidence by 78%, compared with placebo.
No difference in incidence of falls or in metacarpal
bone density, suggesting that the reduced fracture
incidence was due to improved bone quality.
JAMA 2005;293:1082-8
Homocysteine: positive clinical trials
In an uncontrolled trial, supplementation
with folic acid, vitamin B6, and vitamin
B12 appeared to reverse carotid atherosclerosis in hyperhomocysteinemic
patients.
Lancet 1998;351:263.
Homocysteine: positive clinical trials
Meta-analysis of 8 randomized
controlled trials found that folic acid
supplementation significantly reduced
the risk of stroke by 18%.
Lancet 2007;369:1876-1882
Homocysteine: negative clinical trials
Numerous randomized controlled trials have
found that homocysteine-lowering regimens
failed to reduce various endpoints including
all-cause mortality, myocardial infarction,
heart disease-related death, and venous
thrombosis, and failed to slow the
progression of Alzheimer’s disease.
Homocysteine: negative clinical trials
Because many trials have produced
negative results, some researchers have
pronounced the homocysteine theory of
disease “dead.”
Possible explanations for
negative results
Nutrient imbalances from supplementing
with just a few nutrients (e.g.,
exacerbation of magnesium deficiency)
Is postprandial homocysteine more
important than the fasting level?
J Am Coll Cardiol 1985;6:725-30
Possible explanations for
negative results
Betaine (and presumably choline) appear to
be more effective than folic acid, B12, and B6
for lowering postprandial (i.e., post
methionine load) homocysteine levels.
Arterioscler Thromb Vasc Biol 2005;25:379-385.
Potential benefits of betaine and choline
were not investigated in clinical trials.
Betaine and choline:
recommended doses
Betaine: 500 mg/day to 2,000 mg 3 times per
day
Choline: 500-1,000 mg/day
Does folic acid cause cancer?
Double-blind study: 1 mg/day for 3-8 years was
associated with a higher incidence of prostate
cancer (9.7% vs. 3.3% for placebo) in patients with
recent colorectal adenoma. J Natl Cancer Inst 2009;101:432-435
Double-blind JAMA study: 0.8 mg/day was
associated with a significant 21% increase in
cancer incidence, 38% increase in cancer deaths.
JAMA 2009;302:2119-2126
Does folic acid cause cancer?
Study weaknesses
1) Post-hoc analyses of earlier research that was
designed to ask a different question.
2) Studies used folic acid alone or folic acid plus a
few other nutrients. Effect might be different when
part of a comprehensive nutritional program.
3) No increase in cancer incidence in the US since
folic acid fortification of food began in 1998.
JAMA 2009;302:2152-2153
Does folic acid prevent cancer?
Double-blind studies: folic acid supplementation
for 3.0-7.3 years associated with a nonsignificant
decrease in cancer risk in health professionals.
JAMA 2008;300:2012-2021; Am J Clin Nutr 2009;90:1623-1631
Observational studies: Folic acid supplementation
or higher dietary folate intake was associated with
decreased incidence of cancer.
Folic acid prevented cancer in some animal
studies.
Benefits of folic acid
Prevention of neural tube defects
Prevention of strokes (Lancet 2007;369:1876-1882)
Along with B12, prevention of hip fractures in
stroke patients (JAMA 2005;293:1082-1088)
Migraine prophylaxis in patients with elevated
homocysteine levels (Headache 2007;47:1342-1344)
Folic acid and cancer: conclusion
Effect of folic acid on cancer risk and cancer
mortality remains uncertain.
There does not appear to be any compelling
reason to recommend that the general public
avoid folic acid supplementation.
Folate supplements: synthetic vs.
“natural”
Most supplements contain synthetic folic
acid (pteroylglutamic acid; PGA).
PGA is metabolized in vivo to methylfolate,
the form in which the vitamin is normally
transported in the body.
Folate supplements: synthetic vs.
“natural”
Administration of large doses of PGA
(> 400 µg/day?) might lead to the presence
of large amounts of unmetabolized PGA,
which could theoretically have an anti-folate
effect through competitive inhibition of
folate-dependent enzymes.
Folate supplements: synthetic vs.
“natural”
However, aside from that theoretical concern, there
is no obvious reason to believe that synthetic folic
acid is harmful. In addition, there is no evidence
that other commercially available forms of
supplemental folate (such as methylfolate, folinic
acid, or 5-methyltetrahydrofolate) are safer than
PGA.
Folate supplements: synthetic vs.
“natural”
Virtually all of the research demonstrating a
beneficial effect of folate has been done using
synthetic folic acid. Other forms of folate have not
been shown to prevent neural tube defects, strokes,
migraines, or osteoporotic fractures. Synthetic
folic acid would therefore seem to be preferable to
other folate preparations in most circumstances.
Does vitamin A increase mortality?
A review of 5 randomized controlled trials,
which included a total of 21,677
participants, concluded that vitamin A
supplementation increased the risk of death
significantly by 16%, compared with
placebo.
JAMA 2007;297:842-857.
Does vitamin A increase mortality?
Three of those trials, which included
96% of the pooled participants, are
described in the following slides.
First vitamin A study
N = 18,314 (84.5% of the pooled participants).
Participants were smokers, ex-smokers, or workers
exposed to asbestos.
Intervention: Placebo or beta-carotene plus vitamin
A. Beta-carotene has been shown to cause lung
cancer in this population group.
N Engl J Med 1996;334:1150-1155
First vitamin A study
Even if vitamin A did contribute to the increase in
mortality, the effect may not apply to non-smokers.
There might be a toxic interaction between vitamin
A and tobacco smoke, analogous to the toxic
interaction demonstrated between vitamin A and
chronic alcohol consumption.
Alcohol Clin Exp Res 1983;7:15-21
Second vitamin A study
N = 2,297 (mean age, 63 years; 10.6% of pooled
participants). 25,000 IU/day or placebo for ≤ 5 years.
Death rate during the study: vitamin A, 2.07%;
placebo, 2.11%
Death rate including treatment period plus posttreatment follow-up: vitamin A, 5.4%; placebo, 4.6%
25,000 IU/day may be too high for some older people.
Findings may not apply to general population.
Third vitamin A study
N = 109. Elderly nursing home residents: 200,000 IU
of vitamin A (single dose) or placebo. During followup, death rate higher with vitamin A than with placebo
(11.3% vs. 7.1%)
Vitamin A group on average of 4.8 years older than
placebo group.
Should we be giving frail old people such a massive
dose of vitamin A?
The other two “vitamin A” studies
Elderly people received a multivitaminmultimineral preparation containing 16 or
17 components (including a small dose of
vitamin A (2,667 IU/day). No significant
difference in mortality between groups.
Does vitamin A cause osteoporosis?
Observational studies: higher intake of
vitamin A or higher serum vitamin A levels
were associated with lower bone mineral
density (BMD) or increased fracture risk.
Adverse effect seen even at low intake levels
(≥ 6,667 IU/day; RDA = 2,333 IU/day).
Does vitamin A cause osteoporosis?
Other observational studies: no association
between vitamin A intake or serum vitamin A
levels and BMD or fracture risk.
One study: higher serum vitamin A levels
were associated with a nonsignificant
decrease in fracture risk.
J Bone Miner Res 2005;20:913-920
Does vitamin A cause osteoporosis?
Vitamin A at a human-equivalent dosage of
2.4 million IU/day caused bone abnormalities in rats, but the equivalent of 470,000
IU/day had no adverse effect. Bone 2003;31:685-689
Healthy men: 25,000 IU/day for 6 weeks had
no effect on serum markers of bone turnover.
J Nutr 2002;132:1169-1172
Observational studies:
confounding factors
Main dietary sources of vitamin A, aside
from liver, are fortified breakfast cereals
(usually with added sugar), fortified
milk, and fortified margarine. Higher
vitamin A intake may simply be a
marker of increased consumption of
these foods.
Diet and osteoporosis
Refined sugar: Adverse effect according to
animal studies and observational studies.
Milk: effect unclear. Associated with increased
fracture incidence on one study.
Am J Public Health 1997;87:992-997
Margarine: Butter contains vitamin K2.
Margarine contains hydrogenated vitamin K2
(dihydrophylloquinine), which is inactive.
Diet and osteoporosis
Liver: Accumulates lead and cadmium (both
of which can cause osteoporosis). Also may
accumulate various xenobiotic chemicals that
could promote bone loss by inhibiting
androgen activity.
J Endocrinol 1998;158:327-338
Strontium and bone:
to dose or to megadose?
Typical diet provides 1-3 mg/day of strontium
Significant amounts lost in refining of flour
Strontium has high affinity for bone; promotes
mineralization of bones and teeth
Stimulates bone formation, inhibits bone
resorption
Two-year strontium clinical trial
Strontium
dose
% change in
lumbar BMD
Incidence of new
vertebral fractures
Placebo
+0.50
54.7%
170 mg/day
+1.35%
38.8%
340 mg/day
+1.65%
56.7%
680 mg/day
+2.97%
42.0%
Study of 353 postmenopausal women with osteoporosis and a history of at least
one vertebral fracture. J Clin Endocrinol Metab 2002;87:2060-2066.
Three-year strontium clinical trial
1,649 postmenopausal women, 680 mg/day of strontium for 3 years. Top
half: % increase in BMD. Bottom half: % reduction in fracture risk.
Adapted from N Engl J Med 2004;350:459-68.
Potential adverse effects
of high-dose strontium on bone
• Syndrome resembling rickets in animals fed 1.5-3.0%
strontium
• Bone mineralization defects in young rats at diet
concentrations of 0.19% or greater in (equivalent to
approximately 800 mg/day for humans). Rat diet
contained 0.5% calcium. Bone 1990;11:313-9.
• High soil strontium concentrations associated with
increased prevalence of rickets in Turkish children
Arch Dis Child 1996;75:524-6.
Other potential adverse effects
of high-dose strontium
• Increased thyroid weight in rats fed 395 ppm of
strontium
• Decreased pituitary weight in rats fed 98.7 ppm or 1,580
ppm, but not 395 ppm.
• Estimated no-observed-adverse-effect level = 98.7 ppm,
equivalent to 41.1 mg/day for humans (calculation based
on 2,000 kcal/day, 30% fat = 417 g/day of food, dry
weight)
Toxicology 1977;7:11-21.
Adverse effects of strontium in
3-year clinical trial
Elevated CPK in 3.4% of patients receiving
strontium, 1.8% of those receiving placebo.
Elevations usually transient.
No mineralization defects found, but only mature
(lamellar) bone was biopsied, whereas adverse
effects would presumably be most pronounced in
new bone.
N Engl J Med 2004;350:459-468.
Three-year strontium clinical trial
1,649 postmenopausal women, 680 mg/day of strontium for 3 years. Top
half: % increase in BMD. Bottom half: % reduction in fracture risk.
Adapted from N Engl J Med 2004;350:459-68.
Two-year strontium clinical trial
Strontium
dose
% change in
lumbar BMD
Incidence of new
vertebral fractures
Placebo
+0.50
54.7%
170 mg/day
+1.35%
38.8%
340 mg/day
+1.65%
56.7%
680 mg/day
+2.97%
42.0%
Study of 353 postmenopausal women with osteoporosis and a history of at least
one vertebral fracture. J Clin Endocrinol Metab 2002;87:2060-2066.
Distribution of strontium in bone
At high doses, most strontium is incorporated by
exchange onto the crystal surface. This strontium, which
may promote bone formation and inhibit bone resorption,
is rapidly lost from bone and excreted in the urine when
supplementation is stopped.
A few strontium atoms are incorporated into the crystal
lattice; this strontium may enhance bone quality, and
appears to persist in bone after supplementation is
stopped.
Bone 2001;28:446-453
Strontium and bone: my opinion
For established osteoporosis, high-dose strontium
(170-680 mg/day) appears to be appropriate for 1-3
years. Thereafter, consider “nutritional” doses
(such as 2-6 mg/day) for longer-term use. Longterm clinical trials (> 3-5 years) needed to determine safety and efficacy of high-dose strontium.
For osteoporosis prevention, “nutritional” doses
may be most appropriate.