|
|
Report Regarding the Proposed Health Claim Concerning 0.8 mg of Folic Acid
Michael J. Glade, Ph.D.
The attached report addresses scientific issues relating to the statement, ".8 mg of folic acid in a
dietary supplement is more effective in reducing the risk of neural tube defects than a lower amount in foods
in common form." Specific issues that have been addressed include:
 current intakes of dietary folates in the United States
bioequivalence of food folates and folic acid
maternal dietary folate intake required to reduce the incidence of fetal
neural tube defects
maternal folic acid intake required to reduce the incidence of fetal neural
tube defects
maternal red blood cell folate concentration as an indicator of risk for fetal
neural tube defects
red blood cell folate concentration as an indicator of risk for cancer,
maternal dietary folate intake required to increase maternal red blood cell
folate concentration
maternal folic acid intake required to increase maternal red blood cell folate
concentration
maternal plasma folate concentration as an indicator of risk for fetal neural
tube defects
maternal plasma total homocysteine concentration as an indicator of risk for
fetal neural tube defects
folic acid intake required to decrease plasma total homocysteine concentration
safety of dietary folates
safety of folic acid
homocysteine and cardiovascular disease
Current Intakes of Dietary Folates in the United States
It is estimated that the average adult in the US consumes about 200 mcg/day of naturally-occurring food
folates.1-5 Although it is estimated that fortification of grains with folic adds on average another
100 mcg/day of folic acid,6 it has been calculated that this additional folic acid intakes increases by
only 3% the proportion of women who are consuming 400 mcg/day of food folate equivalents.7
Bioequivalence of Food Folates and Folic Acid
The naturally-occurring food folates consist of a mixture of conjugates of pteroylpolyglutamic acid which
vary in the number of glutamate residues present. Synthetic folic acid (pteroylglutamic acid), preferred in
dietary supplements and food fortification programs because of its much greater stability, contains a single
glutamate residue.
The available data indicates that food folates and synthetic folic acid are far from bioequivalent.
The bioavailability of food folates generally is estimated to be about half that of folic acid.
8,9,10,11,12,13,14,15 It has been reported that 1 mcg of food folates is approximately
bioequivalent to 0.6 mcg of folic acid when the folic acid is consumed as part of either fortified foods
or dietary supplements consumed with food.16,17 The bioequivalence of food folates is even
lower (about 50%) when compared to folic acid ingested as a dietary supplement between meals.16,17
Despite these apparent differences, folic acid consumed as part of fortified foods and folic
acid in dietary supplements generally are considered to be of similar bioavailability.12,18
However, the bioavailability of folic acid consumed as part of fortified foods may be much lower than has
been assumed; the relative bioavailabilities of folic acid consumed as part of fortified bread, rice and
maize have been reported to be only 30%, 60% and 50%, respectively, when compared to the bioavailability of
folic acid in dietary supplements.19,20,21
On an equal weight basis (the most commonly used basis for comparison in the scientific literature), folic
acid has been found to be more effective than mixed food folates in increasing plasma and red blood cell
folate concentrations12,22 and improving folate status.12 In one study,
supplementation with 400 mcg/day of folic acid induced significant increases in red blood cell folate
concentrations, while the same amount of supplemental food folates did not.12 Furthermore,
in a study of the impact of food folates and folic acid on the incidence of first occurrences of neural tube
defects, high maternal intake of food folates was only half as effective as maternal folic acid
supplementation.23
Of additional concern are the observations that beans and some other vegetables contain factors that inhibit
the intestinal absorption of dietary folates24 and that about 20% of food folates are
destroyed by cooking25,26,27 while another 50% to 80% of food folate content may be lost
following post-harvest exposure to light.28 The effects of cooking are especially
troublesome because the best sources of dietary folates (liver, kidney, spinach, asparagus, broccoli,
soybeans) are unlikely to be eaten raw.
Maternal Dietary Folate Intake Required to Reduce the Incidence of Fetal Neural Tube Defects
The use of maternal diet and oral supplementation to reduce the incidence of first occurrences of fetal
neural tube defects and their recurrence has been studied for over two decades. Investigators have examined
the results of historic consumption of dietary folates as well as the consumption of folic acid in a wide
range of daily intakes as a component of "multivitamins" and as the sole ingredient of dietary
supplements.
Two studies relating maternal dietary folate intakes to the incidence of fetal neural tube defects have been
retrospective case-control studies, in which the outcome of the variable of interest (presence or absence of
neural tube defects) is known before intakes are estimated. This can bias in subtle ways the determination
of intake intervals upon which comparisons are then based, often in favor of "discovering" a treatment
effect, and decreases the reliability of conclusions drawn from such analyses. In one such study, estimated
maternal consumption of more than 300 mcg/day of food folates was associated with a significantly reduced
incidence of first occurrences of fetal neural tube defects, compared to maternal consumption of less than
200 mcg/day of food folates (odds ratio: 0.6).23 Other investigators observed that when the
outcomes among 1100 pregnancies were compared between mothers in the highest quartile of food folate
consumption and those in the lowest, the odds ratio for first occurrences of fetal neural tube defects was
0.69 (95%n confidence interval: 0.47 – 1.0).29
In a retrospective cohort study of 12,000 pregnancies, suffering similar limitations to the case-control
studies, the relative risk for first occurrences of fetal neural tube defects in mothers consuming more than
100 mcg/day of food folates, compared to mothers consuming less than 100 mcg/day of food folates, was 0.42
(95% confidence interval: 0.16 – 1.15).30
The available evidence indicates that the potential for maternal dietary folate consumption, in amounts
ranging from somewhat greater to much greater than current average intakes, to reduce the incidence of first
occurrences of fetal neural tube defects remains uncertain.
Maternal Folic Acid Intake Required to Reduce the Incidence of Fetal Neural Tube Defects
Five randomized double-blind placebo-controlled prospective ("gold-standard") trials have examined
the hypothesis that maternal periconceptional supplementation with folic acid could reduce the incidence
of first occurrence of fetal neural tube defects or their recurrence. Several other experimental designs
(nonrandomized controlled trial, retrospective observational case-control study, retrospective cohort study)
also have been used to test this hypothesis.
In a landmark study, the incidence of fetal neural tube defect recurrence was reduced 72% in 593 women who
consumed a folic acid supplement providing 4000 mcg/day, compared to the incidence in 602 women who received
a placebo (relative risk for fetal neural tube defect recurrence: 0.28; 95% confidence interval: 0.12 – 0.71)
.31 This result is similar to that of an older randomized double-blind placebo-controlled
prospective trial, in which 4000 mcg/day of maternal periconceptional folic acid supplementation reduced
the incidence of recurrent fetal neural tube defects by 60%.32 In a somewhat less
convincing nonrandomized controlled study, an experimental design that would be expected to favor treatment
effects, 360 mcg/day of maternal periconceptional folic acid supplementation reduced the recurrence of fetal
neural tube defects by 86%.33
In contrast, the results of two randomized double-blind placebo-controlled prospective trials suggest that
supplemental maternal folic acid intakes of 400 mcg/day34 and 5000 mcg/day35 had no effect on the recurrence
of fetal neural tube defects. However, in those studies the incidences of fetal neural tube defects among
the babies of unsupplemented mothers were only 1%34 or 4%,35 suggesting that neither trial has sufficient
power to detect a protective effect of supplemental folic acid even if such an effect had occurred.
In a pivotal randomized double-blind placebo-controlled prospective study, the incidence of first occurrence
of fetal neural tube defects was reduced 70% by maternal periconceptional supplementation with 800 mcg/day
of folic acid.36 Positive preventive effects on the occurrence of first fetal neural tube defects also have
been observed in a series of retrospective observational case-control studies, in which maternal
periconceptional supplementation with 400 mcg/day to 1800 mcg/day of folic acid was associated with 30% to
70% reductions in the incidence of first fetal neural tube defects.23,29,30,34,37,38 However, it is not
possible from these reports to ascertain the nature of the dose-dependency of the degree of risk
reduction.
The results of a case-control study have been interpreted to suggest that maternal periconceptional
supplementation with 800 mcg/day of folic acid failed to reduce the incidence of first fetal neutral
tube defects in that study.39 However, other investigators have suggested that neural
tube defects were defined incorrectly in that study, placing the results and their interpretation in doubt.
40
In addition to reduction in the risk of first fetal neural tube defects, maternal periconceptional
supplementation with 800 mcg/day of folic acid also was reported to reduce by about 50% the incidence of
major non-neural tube congenital abnormalities, including heart defects, conotruncal defects, renal tract
anomalies, obstructive urinary tract anomalies, limb defects and congenital hypertrophic pyloric
stenosis.41 Similarly, maternal periconceptional supplementation with "multivitamins
" providing at least 400 mcg/day of folic acid has been associated with a 64% reduction in transposition
of the great arteries in the newborn,42 59% reduction in the incidence of isolated
conotruncal defects,42 85% reduction in the incidence of neonatal urinary tract
anomalies,43 25% to 60% decrease in the incidence of orofacial clefts,44,45
35% decrease in the incidence of neonatal limb defects,46 and an approximately 50% reduction
in the incidence of premature parturition and intrauterine growth retardation.47
After reviewing much of the same evidence presented here, the Centers for Disease Control and
Prevention (CDC) concluded in 1992 that "supplementation with a multivitamin containing 400 mcg of folic
acid prevents the occurrence of over 50% of neural tube defects when it is taken before conception and
continued throughout the fist trimester of pregnancy."48 A subsequent CDC analysis
suggested that increasing average maternal periconceptional folic acid intake (as supplements, fortified
foods or both) to 500 mcg/day could triple the number of prevented fetal neural tube defects.49
Consequently, a more recent (1996) CDC recommendation is for maternal periconceptional
supplementation with 250 mcg/day of folic acid in addition to dietary intake of at least 200 mcg/day of
food folates,50 equivalent to 350 mcg/day of supplemental folic acid (or 700 mcg/day of
food folates). These intake levels also have been recommended by the American College of Preventive
Medicine.51
In August, 1999, the American Academy of Pediatrics recommended that "all women capable of becoming
pregnant consume 400 mcg of folic acid daily."7 This level of supplementation also is
recommended by the Irish Department of Health,52 the Department of Health of the United
Kingdom,53 the Dutch Health Council,54 the Canadian Task Force on the
Periodic Health Examination,55 the National Health and Medical Research Council of Australia,
56 and the Scottish Office Home and Health Department.57
After holding public hearings and soliciting scientific input from a variety of sources, the Institute of
Medicine recommended in 1998 that "all women capable of becoming pregnant should consume 400 mcg of
synthetic folic acid from fortified foods and supplements in addition to intake of food folate from a varied
diet." If average dietary folate intake in the US is about 200 mcg/day,1-5 this
recommendation is equivalent to about 1000 mcg/day of naturally-occurring dietary food folates, or at least
500 mcg/day of synthetic folic acid (assuming 50% bioavailability of dietary food folates).
Others have asserted that a recommended periconceptional folic acid intake of 400 mcg/day appears to
represent an unscientific compromise of sorts,58 noting that 400 mcg/day of folic acid is
expected to decrease the incidence of fetal neural tube defects by only about half,49 when the most rigorous
evidence obtainable (that from a "gold-standard" randomized double-blind placebo-controlled
prospective trial36) indicates that much greater protection can be obtained from a daily
periconceptional folic acid intake of 800 mcg/day68 and the optimal intake of folic acid
appears to be between 400 mcg/day and 800 mcg/day and may well be 800 mcg/day.69,70,71
Noting that folate metabolism is accelerated during gestation,72 others have recommended
daily folic acid intakes of at least 400 mcg during the first trimester, 660 mcg during the second trimester
and 470 mcg during the third trimester.73,74,75
Maternal Red Blood Cell Folate Concentration as an Indicator of Risk for Fetal Neural Tube Defects
"Normal" red blood cell folate concentrations range between 200 mcg/L and 300 mcg/L.76
On average, red blood cell concentrations have been found to be lower in the mothers of infants
with neural tube defects than in mothers of unaffected babies.77,78 Red blood cell folate
concentrations have been reported to be below the standard normal range in 27% of women giving birth to
infants with neural tube defects16 and to be inversely proportional to the risk for fetal
neural tube defects.79,80,77,78,81,82,83,84,85,86 Maternal red blood cell folate
concentrations greater than 400 mcg/L have been associated with a relative risk for fetal neural tube defects
of 0.12, compared to the risk when maternal red blood cell folate concentrations are less then 150 mcg/L.
87,81
Red Blood Cell Folate Concentration as an Indicator of Risk for Cancer
Red blood cell concentrations below 200 mcg/L in adult men and women have been associated with increased
risk for the development of colorectal cancer,88 colorectal adenoma,89
colonic adenoma,90 progression of ulcerative colitis to dysplasia and colon cancer,91
gastric cancer,92 esophageal cancer,93 cervical dysplasia,
94 and bronchial squamous metaplasia, especially in cigarette smokers.95 In
contrast, risk for colorectal cancer96,97 colorectal adenoma89,98,99 and
progression of ulcerative colitis to colon cancer100,101 are inversely proportional to
dietary folate intake. It has been estimated that every 10 mcg/L increase in red blood cell folate
concentration decreases the risk for colon cancer by 18%91 (possibly by fostering increased
methylation of colonocyte DNA102).
Maternal Dietary Folate Intake Required to Increase Maternal Red Blood Cell Folate Concentration
Food folate intakes of 400 mcg/day have been reported to have little effect on red blood cell folate
concentrations.26
Maternal Folic Acid Intake Required to Increase Maternal Red Blood Cell Folate Concentration
Red blood cell folate concentration has been shown to be directly proportional to supplemental folic acid
intake.67,103,104,105 Although supplemental folic acid intake of 400 mcg/day significantly
increases red blood cell folate concentration,26 this level of intake does not saturate red
blood cell folate content.67 Greater daily intakes of folic acid are able to further load
red blood cells.67
Maternal Plasma Folate Concentration as an Indicator of Risk for Fetal Neural Tube Defects
Folate in plasma is labile and plasma folate concentrations may reflect transient changes in folate status,
while red blood cell folate concentrations are less subject to short-term variation and may more accurately
reflect physiologic folate status.76 Nonetheless, maternal plasma folate concentrations
below 5 mcg/L have been associated with increased risk for fetal neural tube defects106,107
and early pregnancy loss.108,109 "High normal" (greater than 12 mcg/L) plasma
folate concentrations have been reported to decrease risk for coronary artery disease110,111,112,113
and cervical dysplasia.114 Conversely, plasma folate concentrations below
5 mcg/L have been reported to increase risk for coronary artery disease115 and ischemic
stroke.116
Maternal Plasma Total Homocysteine Concentration as an Indicator of Risk for Fetal Neural Tube
Defects
Because complex interconversions can confound measurements of homocysteine in plasma, most investigators in
fact measure "total homocysteine" in plasma samples; this measurement will include the sum of free
homocysteine, cysteine-homocysteine mixed disulfide, and protein-bound homocysteine in plasma.
Although there is some disagreement, "elevation" in plasma total homocysteine (tHcy) concentration
is taken to begin at concentrations of 13.6 micromol/L,117 14 micromol/L,118
15 micromol/L119 or 16.3 micromol/L.120
Many women who give birth to babies with neural tube defects have elevated concentrations of tHcy in
plasma106,121 and amniotic fluid.122,123,124 On average, plasma tHcy
concentrations are greater in mothers of affected babies than in mothers of unaffected babies.
125,126,127,128,129,107,130 Elevated plasma tHcy concentration in the mother increases the
odds ratio for fetal neural tube defects.131 Mothers in the highest quartile of plasma
tHcy concentration have the greatest risk for fetal neural tube defects.132 On average,
plasma tHcy concentrations are higher in children with spina bifida than in children without neural tube
defects.133 Pregnant women with classic homocysteinuria suffer 50% fetal loss.134
Similarly, it has been reported that 20% of women with recurrent early pregnancy loss had
elevated plasma tHcy concentrations,108,109 18% of women with a history of severe
early-onset pre-eclampsia had a decreased rate of conversion of homocysteine to methionine and elevated
plasma tHcy concentrations,135 and women with a history of placental abruption or infarction
had a 3-fold greater incidence of decreased rate of conversion of homocysteine to methionine.136,137
When exposed experimentally to homocysteine in concentrations approximating those found in the
plasma of women who have given birth to infants with neural tube defects, avian embryos develop neural tube
defects.138
Folic Acid Intake Required to Decrease Plasma Total Homocysteine Concentration
Although plasma tHcy concentrations tend to be inversely proportional to red blood cell folate
concentrations,108,109,8,139 and increasing folic acid intake by 100 mcg/day (the increment
expected to result from the current level of folic acid fortification in the US food supply140,141
) has been shown to increase red blood cell folate concentrations by about 20%,105
this increase in folic acid intake has not been found to reduce plasma tHcy concentrations.142
Similarly, supplementation with 170 mcg/day of folic acid had no effect on average plasma tHcy
concentration in healthy nonpregnant women of childbearing age with initial plasma tHcy concentrations
below 15 micromol/L; in contrast, daily supplementation with either 270 mcg or 370 mcg of folic acid
reduced plasma tHcy concentration by an average of 35%.143 Consistent with these reports
is the finding that either 100 mcg or 200 mcg of daily supplemental folic acid had no effect on initially
elevated plasma tHcy concentrations in male hospital workers; in contrast, 400 mcg/day of folic acid reduced
plasma tHcy concentrations by over 50%.144
At least 200 mcg/day of supplemental folic acid
is required to accelerate the conversion of homocysteine to methionine145 and both reduce plasma tHcy
concentrations from greater than 15 micromol/L to less than 15 micromol/L146 and maintain
this lower concentration with no other changes in diet or lifestyle.48 Similarly, the
results of a meta-analysis suggest that 200 mcg/day of supplemental folic acid may reduce plasma tHcy
concentrations by an average of 4 micromol/L;147 if this effect is linear at elevated
concentrations of plasma tHcy, then adults with initial plasma tHcy concentrations of 30 micromol/L
(the approximate average among several groups of adults with initially elevated plasma tHcy concentrations
148,149,150) would require 800 mcg/day of supplemental folic acid to reduce their plasma
tHcy concentrations to below 15 micromol/L.
This estimate is consistent with a report that daily supplementation with 650 mcg of folic acid was
associated with, on average, a 42% decrease in plasma tHcy concentrations in 100 adult men and women.
151 In other studies of adult men and women, daily supplementation with 400, 1000 or 5000 mcg of
folic acid was associated with approximately 50% or greater decreases in average plasma tHcy concentrations.
152,148,149,153 These reports have led to the conclusion that daily supplementation with
650 mcg of folic acid is required for "normalization"of plasma tHcy concentration.48
Safety of Dietary Folates
There are believed to be no concerns limiting dietary folate intake. Indeed, intakes sufficiently high to
risk interfering with the diagnosis of pernicious anemia (a persistent red herring inappropriately attached
to safety evaluations of folic acid) can only be obtained by eating large amounts of foods that also supply
plentiful amounts of vitamins B6 and B12.154 For example, typical Puerto Rican diets
provide 1000 to 2000 mcg/day of dietary folates, with no adverse effects.155
Safety of Folic Acid
Folic acid has no known toxicity in humans.7 Intakes of 10000 mcg/day for 5 years have had no adverse
effects,156,157 although intakes greater than 1000 mcg/day may reduce the effectiveness of
some anticonvulsant medications.(158,159 Folic acid supplements taken by women have not
caused long-term developmental, behavioral or neurologic effects in their children.160
Contrary to an earlier speculation, folic acid supplementation does not decrease the incidence of fetal
neural tube defects by increasing the rate of fetal death158 or by increasing the risk for
miscarriage (relative risk: 1.06; 95% confidence interval: 0.79 – 1.43).161
Concern has been expressed over the theoretical possibility that substantial supplementation with folic acid
could resolve the folate-sensitive anemia of vitamin B12 deficiency (by supplying methyl groups for DNA
replication and methylation, eliminating the need for vitamin B12-dependent recycling of methyl group donors)
while allowing the folate-insensitive demyelination neuropathy of vitamin B12 deficiency to progress
undiagnosed and unchecked. It has been speculated that by removing easily recognized pernicious anemia from
the diagnostician's repertoire, widespread folic acid supplementation or food fortification with folic acid
at biologically optimal levels might trigger an epidemic of undiagnosed and irreversible neuropathy.
Despite the frequency and sometimes ardor with which this objection to maximization of the benefits of folic
acid supplementation and food fortification has been voiced, it has never been demonstrated that the
diagnosis of vitamin B12 deficiency is in fact made more promptly in cases presenting with anemia; it also
has yet to be demonstrated that folic acid hastens the progression of the neuropathy of vitamin B12
deficiency.162 After a decade of lack of scientific support, the theoretical possibility
that folic acid supplementation or food fortification could interfere with the recognition, diagnosis and
treatment of vitamin B12 deficiency should be abandoned as a public health concern.
Homocysteine and Cardiovascular Disease
Homocysteine has become recognized as an independent risk factor for atherosclerosis
163,164,165,166,167,168,169,170,171 and coronary artery disease.169,172,173,174
The risks for atherosclerosis175,166,169,176,177,178,179 and the severity of
carotid artery stenosis180,181 are proportional to plasma tHcy concentration. In addition,
plasma tHcy concentrations greater than 15 micromol/L increase the risk for peripheral vascular disease,
174 stroke,174,182 myocardial infarction,174,183,184
cerebral infarction,185 carotid artery stenosis,180 isolated systolic
hypertension,186 occurrent thrombosis,187,188 recurrent thrombosis,
187,188,189 venous and arterial occlusive disease,190,191 intermittent
claudication,192 progression of lower extremity peripheral artery disease,193
diabetic retinopathy,194 and elevated serum low-density lipoprotein-cholesterol
concentration.174 Elevated plasma tHcy concentrations have been observed in 28% to 38% of
patients with peripheral artery disease,196,197 23% of patients with intermittent
claudication,195 25% of patients with recurrent venous thrombosis,189 30%
of patients with acute strokes and transient ischemic attacks,198 and 40% of stroke
survivors.198
It has been estimated that the risk for coronary artery disease increases in men by 60% and in women by 80%
for every 5 micromol/L increase in plasma tHcy concentration above 15 micromol/L.169,173,199
This increase in risk is similar to that associated with an increase of 20 mg/dL in serum cholesterol
concentration.173 The magnitude of the risk for coronary artery disease associated with a
plasma tHcy concentrations greater than 15 micromol/L has been equated to the risk associated with cigarette
smoking.169,172
Although the mechanisms by which homocysteine contributes to such an array of cardiovascular diseases
remain uncertain, they have received much attention. For example, homocysteine has been reported to
directly inhibit endothelial production of nitric oxide and prostacyclin while stimulating release of
vasoconstriction-promoting prostaglandins.123 Homocysteine also appears to induce
activation of endothelial factor V,200,201 stimulate thromboxane formation and the binding
of lipoprotein(a) to fibrin,202 and inhibit thrombomodulin activity200,201
and fibrinolysis.202 Induction of hypercoagulability, along with stimulation of arterial
smooth muscle cell proliferation203 and facilitation of LDL oxidation,204,205
could combine with increased tendency to vasoconstriction to promote vascular blockage.209
Furthermore, homocysteine is directly toxic to the vascular endothelium,206,207,208
inducing endothelial cell desquamation and the cell-mediated decomposition of basement membrane
proteoglycans.202 The sensitivity of the vascular endothelium to homocysteine may be
explained by the observation that human heart, vascular tissue and endothelial cells lack the enzymes that
convert homocysteine to cysteine, prolonging their exposure to this toxic metabolite of methionine
metabolism.
Conclusions
1. The majority of adult men and women in the US consume less than the 1989 RDA for dietary folate
(180-200 mcg/day).
2. Fortification of food with folic acid at current levels (140 mcg of folic acid per 100 g of cereal grains)
has little if any impact on folate status in the US.
3. The rigorous evidence obtainable (that from a "gold-standard" randomized double-blind placebo-
controlled prospective trial) concerning the relationship between maternal periconceptional folic acid intake
and the incidence of fetal developmental abnormalities (including neural tube defects, heart defects,
conotruncal defects, renal tract anomalies, obstructive urinary tract anomalies, limb defects and congenital
hypertrophic pyloric stenosis), data concerning folate metabolism during pregnancy, and studies of the
relationship between folic acid and homocysteine, consistently indicate that protection is not maximized by
daily periconceptional folic acid intakes of less than 650 mcg/day; 800 mcg/day may be optimal.
4. Maternal daily periconceptional intake of 300 mcg or more of dietary folates has had less impact on the
incidence of first occurrences of fetal neural tube defects than has maternal periconceptional
supplementation with 800 mcg/day of folic acid.
5. There are no safety issues limiting dietary folate intake.
6. Folic acid has no known toxicity in humans and has not been demonstrated to "mask" vitamin B12
deficiency.
|
