Determination of the vitamin B-6 requirement of pregnant women


Material Information

Determination of the vitamin B-6 requirement of pregnant women
Physical Description:
vii, 141 leaves : ill. ; 28 cm.
Schuster, Karen Ann, 1950-
Publication Date:


Subjects / Keywords:
Vitamin B6 in human nutrition   ( lcsh )
Pregnancy -- Nutritional aspects   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1983.
Includes bibliographical references (leaves 119-140).
Statement of Responsibility:
by Karen Schuster.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000381726
notis - ACC2257
oclc - 10328070
System ID:

This item is only available as the following downloads:

Full Text







To Glenn and his father Gordon,
for their love of learning


The author gratefully acknowledges the assistance,

encouragement and support provided by her major advisor, Dr.

Lynn B. Bailey, during the research and writing of this

dissertation. The author also thanks the members of the

supervisory committee, Dr. James S. Dinning, Dr. Jesse F.

Gregory III, Dr. Charles S. Mahan and Dr. Harry S. Sitren,

for their valuable advice and guidance.

For making this research possible, the cooperation of

the Maternity and Infant Care clinics personnel,

nurse-midwives, and clients is sincerely appreciated.

Gratitude is also extended to those students, staff and

faculty who participated in the nonpregnant portion of the

study. The author is also very grateful to the

Hofmann-LaRoche Company for generously providing the vitamin

B-6 supplements for this study.



ACKNOWLESGMENTS ....................................

ABSTRAC T I.. .......... ................* ...* *

LITERATURE REVIEW ................................ 4

Vitamir B-6 History ............... .........
Vitamin B-6 Chemistry, Absorption and
B :cavailati t ............................
Vitamin B-6 Metabolism...................... 10
Vitamin B-6 Requirements.................. .. 15
Vitamin B-6 Deficiency ........ .......... 21
Methods of Vitamin B-6 Status Assessment
and Vitamin B-6 Analys s ............ ....... 30
Vitamin B-6 Status and Requirement
During Preg ancy .................. ....... *

EXPERIMENTAL PROCEDURES............... .......... 48

Subjects ...................................... 48
Experimental Desigr ...........................-
Research P? otorc ............................. 5
Sample Colle tion............................. 52
Bicc emi al Analyses ...................... ..... 53
Dietary Analysis............. ...... .... 64
Stat:stical Anail s:s .......................... 64

RESULTS ............. ............ ............ 67

Nc- e re a r t o u ....................... .. .... 67
Fre gnatr' 3 p.................. ............... 69

D ISCU SS O C! ......................................... 95

Nutr :et 1 nta ke..............................
Biocnum:cal: I:-.' atcrs : Vitamin B-6 Status.. 96
Vitamin B-6 Statws a I-r tial Visit ........... 98
Ef-e:t := 7;tam:n -6 Su:==:eserntatoCr
on V:ta-i" B-6 Status.................... 100

Effect of Vitamin B-6 Supplementation and
Status on Birth Outcome.................... 103
Vitamin B-6 Requirement during Pregnancy...... 105
Vitamin B-6 Status and Morning Sickness....... 108
Vitamin B-6 and Pre-eclampsia................. 109

CON CLU SION S ........................................ ii l

APPEND IX ........... ....................... ....... 14
F orm 1................ ....................... 114
Form 2 ........................................ 116

REFERENCES ......................................... 119

BIOGRAPHICAL SKETCH ................................ 141

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Karen Schuster

August 1983

Chairman: Dr. Lynn B. Bailey
Major Department: Food Science and Human Nutrition

The objective of this study was to determine the

vitamin B-6 requirement of pregnant women by assessing the

effect of graded doses of maternal pyridoxine

supplementation on the biochemical indicators of vitamin B-6

status and on the condition of their infants at birth. The

196 volunteer subjects were low-income patients attending

Maternal and Infant Care (MIC) clinics in north central

Florida and ranged in age from 17 to 38 years. The mean

stage of pregnancy at the initial clinic visit was 15+4

weeks and ranged from 6 to 21 weeks. At the first prenatal

clinic visit subjects were randomly assigned a daily vitamin

B-6 supplement containing 0, 2.6, 5, 7.5, 10, 12.5, 15 or 20

mg of pyridoxine-HCl. Maternal health and vitamin B-6 status

of the subjects were assessed at three stages of pregnancy:

first prenatal clinic visit, 30 weeks gestation, and at

term. Vitamin B-6 status and condition of the infants at

birth were also determined.

The mean dietary vitamin B-6 intake of the group was

1.43+1.28 mg/day as estimated from 24-hour dietary recalls.

Maternal plasma PLP levels were positively correlated with

vitamin B-6 supplementation at 30 weeks gestation (r=0.55,

p<0.0005) and at delivery (r=0.54, p<0.01). However, cord

plasma PLP levels did not increase linearly with the level

of vitamin B-6 supplementation but reached a maximum at 7.5

mg and greater. Supplemental pyridoxine-HCl at the 5 mg

level was required to maintain plasma PLP at 30 weeks

gestation at a level comparable to initial values; however,

7.5 mg were required to prevent a decrease in maternal

plasma PLP at delivery from the initial level. Apgar scores

at 1 minute after birth were significantly higher (p<0.05)

for infants whose mothers took 7.5 mg or more supplemental

pyridoxine-HCl than for infants of mothers who took 5 mg or

less. These findings indicate that the approximate vitamin

B-6 requirement of pregnant women is between 5.5 and 7.6

mg/day (diet plus supplement as pyridoxine equivalents) to

ensure adequate vitamin B-6 status and optimal health of the

infant at birth.



The 1980 Recommended Dietary Allowance (RDA) for

vitamin B-6 for pregnant women is extrapolated from

depletion-repletion studies with nonpregnant women. The

increase in the RDA during pregnancy from 2 mg to 2.6 mg is

based on the additional vitamin B-6 needed for the increased

protein allowance. No additional vitamin B-6 is recommended

to compensate for fetal demand, increased maternal metabolic

requirements, and hormonal induction of maternal vitamin B-6

dependent enzymes.

Numerous studies have demonstrated that pregnant women

have significantly lower blood vitamin B-6 levels, plasma

pyridoxal 5'-phosphate (PLP) levels, and decreased

erythrocyte aminotransferase activities as well as decreased

PLP saturation of these enzymes compared with nonpregnant

women. Plasma PLP levels of pregnant women have been

reported to be very low at term, with a significant number

having levels indicative of a vitamin B-6 deficiency

relative to nonpregnant women. Several vitamin B-6

supplementation trials have shown that supplementing

pregnant women with 2 to 4 mg of vitamin B-6 is insufficient

to maintain normal biochemical indices of vitamin B-6 status

throughout pregnancy.

These supplementation trials have not controlled for

prior supplementation with vitamin B-6 or oral contraceptive

use, and little attempt to quantify dietary vitamin B-6

intake has been made. These studies have indicated that the

current vitamin B-6 RDA for pregnant women is inadequate to

maintain normal biochemical indices of vitamin B-6 status,

and a carefully designed study with pregnant women receiving

a placebo or graded doses of pyridoxine is needed to

determine the vitamin B-6 requirement of pregnant women.

The objectives of this study were

1. to estimate the requirement for vitamin B-6 of pregnant


2. to examine the effects of graded doses of vitamin B-6

taken by pregnant women on the biochemical indicators of

vitamin B-6 status of the mothers at the 30th week of

pregnancy and at delivery and of the infant at birth,

3. to assess the the effects of graded levels of maternal

vitamin B-6 supplementation on the condition of the infants

at birth,

4. to examine the relationship between vitamin B-6 status

and the degree of morning sickness experienced by pregnant

women during early pregnancy, and

5. to investigate the incidence of gestational diabetes

and pre-eclampsia as related to vitamin B-6 status in a

group of pregnant women.

Pregnant women attending Maternity and Infant Care

clinics in north central Florida were randomly assigned


supplements containing 0 (placebo), 2.6, 5, 7.5, 10, 12.5,

15 or 20 mg of pyridoxine-HCI at their first prenatal

appointment. Maternal health and vitamin B-6 status as

measured by plasma pyridoxal 5'-phosphate levels and

erythrocyte aspartate aminotransferase activity and

stimulation by exogenous PLP were determined at the first

clinic visit, 30 weeks gestation, and at term. The vitamin

B-6 status and condition of the infant at birth were also



Vitamin B-6 History

In 1934 Gyorgy (1) differentiated the "rat pellagra

preventative factor" from riboflavin (vitamin B-2) in rice

bran and called the new vitamin "B-6." Five groups of

scientists independently isolated the crystalline compound

in 1938 (2-6). Its chemical structure was subsequently

characterized, and the vitamin was synthesized by Harris and

Folkers in 1939 and named pyridoxine (7).

The existence of other naturally occurring compounds

which exhibited vitamin B-6 activity was revealed in

microbiological studies (8, 9). In 1944 these forms of

vitamin B-6 were identified as pyridoxal and pyridoxamine by

Snell (10, 11) and synthesized by Harris et al. (12).

Subsequent work indicated that vitamin B-6 functioned as a

coenzyme in the phosphorylated form of pyridoxal (13).

Vitamin B-6 Chemistry, Absorption and Bioavailability

Vitamin B-6 is the recommended generic name for all

3-hydroxy-2-methylpyridine derivatives which exhibit the

same biological activity. These compounds, pyridoxine,

pyridoxal and pyridoxamine and their phosphorylated

derivatives, are shown in figure 1. Occurring as white

.crystals, the B-6 vitamers are water-soluble but less

soluble in alcohol and often insoluble in ether.

Pyridoxamine and pyridoxine hydrochlorides are stable in hot

dilute mineral acid and alkalai, but pyridoxal hydrochloride

is unstable in basic aqueous solutions. All forms are

extremely sensitive to ultraviolet as well as visible light

(14-17) and destroyed by strong oxidizing agents (3, 14).

Vitamin B-6 appears to be absorbed by passive diffusion

as indicated by a linear relationship between vitamin B-6

dosage and urinary excretion in human and animal studies

(18-20). Recent evidence that rat jejunal uptake of

pyridoxine (0.01 IM 10 mM) is directly proportional to its

concentration supports this proposed mechanism (21).

Absorption of tritium-labelled pyridoxine in rats

occurs rapidly from the upper intestine and slightly less

rapidly from the ileum (22). Some absorption from the colon

but none from the stomach was observed when the pyridoxine

was administered directly to those sites. Similar results

were obtained in human studies (23). After uptake by the rat

jejunum, some of the absorbed pyridoxine is phosphorylated

in the intestinal mucosa (24, 25), which then may be

dephosphorylated by a phosphatase enzyme (25). The rates of

uptake and phosphorylation are not altered in vitamin B-6

deficient rats (26).

Studies of the absorption of tritium-labelled pyridoxal

5'-phosphate (PLP) and pyridoxamine 5'-phosphate (PMP)


Vitamin B-6






HO 33 CH20PO3H2 Phosphate derivative

N ) of Vitamin B-6


Pyridoxine 5'-Phosphate

Pyridoxal 5'-Phosphate

Pyridoxamine 5'-Phosphate

Figure 1. Chemical structures and nomenclature of vitamin
B-6 vitamers.









revealed that under normal physiological conditions the

majority was dephosphorylated to pyridoxal and pyridoxamine,

respectively, in the intestinal lumen, and then absorbed by

passive diffusion (27-29). However, direct absorption of PLP

and PMP, albeit at a slower rate, was also demonstrated. In

experiments with everted sacs of rat small intestine,

comparison of the absorption rates of the three forms of

vitamin B-6 revealed quantitative differences (30, 31). The

rate of t r a n s p ort across t h e i n t e s t i n a l

wall decreased in the following order: pyridoxal >

pyridoxamine > pyridoxine, while the rate of uptake in the

jejunal tissue decreased in the order: pyridoxal >

pyridoxine > pyridoxamine which paralleled the order of

phosphorylation of these compounds.

In contrast to the intestinal absorption of vitamin B-6

by passive diffusion, the transport of vitamin B-6 across

the human placenta appears to involve an active mechanism.

This is evidenced by the large positive gradient of total

vitamin B-6 and individual vitamers in the cord blood over

that of the maternal blood (32-34).

The bioavailability and stability of vitamin B-6 in

foodstuffs are important factors in estimating the dietary

intake of the vitamin. The bioavailability and stability of

vitamin B-6 depend upon such factors as food processing,

storage, and diet composition. Pyridoxamine and pyridoxal

predominate in animal tissues while pyridoxine is found in

higher concentrations in plant materials (35). The effect of

thermal processing on vitamin B-6





observed in the early

i950's when infants

a commercially






































agreed when measuring the

vitamin B-6 content





















The stability of vitamin B-6 in




thermal processing.





a 50%

loss of total

vitamin B-6 content
















(simulating breakfast
















degradation which











- -a



* *

room temperature showed a 40% 60% loss of vitamin B-6

activity compared with frozen storage (44).

While loss of vitamin B-6 occurred during retort

processing and storage in low-acid foods such as lima beans

and beef, no such loss took place in the canned tomato juice

concentrate, a high-acid product (39). Little loss of

vitamin B-6 has been observed during storage of vitamin B-6

fortified flour and bread. The stability of pyridoxine-HCl,

the form used to fortify foods, has been reported to be very

good (45, 46). However, the bioavailability of this vitamer

in a fortified rice breakfast cereal has been found to be

low (47).

Tarr et al. (48) estimated the bioavailability of an

"average" mixed American diet to be 71% 79% when compared

with semi-purified formula diet providing pyridoxine-HC1 as

the sole source of vitamin B-6. While vitamin B-6 has been

reported to be fully available in dried beef and lima beans,

the vitamin B-6 in whole wheat flour and non-fat dry milk is

only about 80% available (49). The bioavailability of

vitamin B-6 from whole-wheat bread was 5 10% less than

from bread fortified with pyridoxine-HCl and from white

bread plus a pyridoxine-HCl oral supplement (50).

Intraluminal perfusions of human jejunum revealed that

absorption of vitamin B-6 from orange juice was less than

from synthetic saline and glucose solutions containing

vitamin B-6 (51) which may be due to binding of the vitamin

B-6 in orange juice to a low molecular compound (52).

Recently, the presence of a natural vitamin B-6 derivative,

pyridoxine 5'-P-glucoside has been reported in some cereals

and seeds (53, 54). Among other factors which may influence

vitamin B-6 bioavailability in foods are the presence of

vitamin B-6 antimetabolites or degradation products such as

e -pyridoxyllysine, which may have anti-vitamin B-6 activity

under certain conditions (55-57).

Vitamin B-6 Metabolism

As a cofactor for more than 60 enzyme systems, PLP is

the main metabolically active form of vitamin B-6 although

PMP occasionally also serves as a coenzyme. PLP-dependent

reactions in amino acid metabolism include decarboxylation,

transamination, deamination, racemization and

desulfhydration (58).

Of particular interest in connection with vitamin B-6

is the metabolism of tryptophan, tyrosine, methionine,

cysteine, glutamate, serine, glycine, aspartate and alanine

(59-64). PLP is required for numerous synthetic and

degradation reactions of compounds functioning as

neurotransmitters in the central nervous system including

dopamine, norepinephrine, epinephrine, serotonin, gamma

aminobutyric acid (GABA), taurine and histamine (62, 65,

66). Lipid and carbohydrate metabolic reactions also involve

vitamin B-6 (67-69). PLP plays an important structural role

in the enzyme phosphorylase which catalyzes the breakdown of

glycogen to glucose-1-phosphate (67). The formation of

S-aminolevulinic acid, an intermediate in the synthesis of

heme, requires PLP (70). Vitamin B-6 is involved in the

synthesis of other vitamins and coenzymes. In its very

active role in the tryptophan metabolic pathway, PLP is also

essential to the synthesis of niacin (71). It is

also required for the production of NN 10o methylene

tetrahydrofolate which is necessary for the synthesis of

deoxythymidylate and purines (72). The formation of coenzyme

A from pantothenate also involves PLP (72). Vitamin B-6 also

plays a role in endocrine metabolism and is required in the

synthesis of a variety of hormones (72).

The metabolic interconversions of the vitamin B-6

vitamers are illustrated in figure 2. PLP i-s synthesized by

phosphorylation of pyridoxal by pyridoxal kinase or

phosphorylation of pyridoxine and pyridoxamine followed by

oxidation by pyridoxamine (pyridoxine) 5'-phosphate oxidase.

Dephosphorylation of PLP, PMP and PNP is accomplished by

various phosphatases. Pyridoxal and pyridoxamine and PLP and

PMP can also be interconverted via transamination (73).

Transamination reactions which require PLP as a coenzyme

involve a Schiff base mechanism whereby PLP and PMP are

interconverted (71). Pyridoxal is oxidized to the inactive

metabolite 4-pyridoxic acid by either aldehyde oxidase or an

unspecific HAD+-linked aldehyde dehydrogenase (74). PLP and

PMP are the predominant B-6 vitamers in mammalian tissues

(75, 76), and the PMP:PLP ratio appears to vary with the

type of tissue and vitamin B-6 status.

4-Pyridoxic Acid

Pyridoxine Pyridoxal < Pyridoxamine
1 tA 3 ,

i 4 1 4 1 4
4t 2 < 2 1
Pyridoxine 5'-P > Pyridoxal 5'-P- Pyridoxamine 5'-P

I Pyridoxal kinase

2 Pyridoxamine (pyridoxine) 5'-phosphate

3 Aminotransferases

4 Alkaline phosphatases

5 Aldehyde oxidase and dehydrogenase

Figure 2. General pathway of mammalian vitamin B-6

Glycogen phosphorylase represents the major vitamin B-6

reservoir in the body since about 90% of the vitamin B-6 in

muscle is associated with this enzyme, up to 5% of muscle

protein is glycogen phosphorylase and 40% of the body is

muscle (77-79). Pyridoxal and PLP in the circulation are the

major transport forms of vitamin B-6 in the body. PLP

represents more than 50% of the plasma B-6 vitamers under

normal conditions (80), and most of the PLP is bound to

albumin (81, 82). Plasma PLP clearance rate in man ingesting

a normal diet has been reported to be 31.7 + 2.7 ml/min

(83), which indicates a rapid turnover rate. The turnover

rate of PLP in erythrocytes is much slower (84), and

therefore it represents a different metabolic pool of PLP in

the body. PLP appears to be about equally distributed

between erythrocytes and plasma in the blood of normal

subjects on regular diets (84). PLP has been shown to

interact with hemoglobin (85) and may therefore be bound

principally to hemoglobin in red blood cells (86). Although

free PLP is able to enter red blood cells, PLP completed

with albumin does not (81, 87, 88). Racial differences in

the activity of pyridoxal kinase in erythrocytes have been

reported (89, 90). However, pyridoxal kinase activity does

not appear to be an important regulator of red cell PLP

levels under normal dietary conditions although it does play

a role when pharmacological doses are administered (91).

The liver appears to be a crucial tissue in the

metabolism of vitamin B-6 and has been shown to be the sole

source of plasma PLP (81). The unphosphorylated B-6 vitamers

absorbed by the intestine are rapidly metabolized in the

liver to PLP, most of which are bound to various

intracellular proteins including glycogen phosphorylase and

cytosolic alanine and aspartate aminotransferases (92).

Pyridoxamine (pyridoxine) phosphate oxidase activity, which

is regulated in vivo by product inhibition, is one factor in

the control of synthesis of liver PLP (93). Under normal

conditions most of the newly synthesized PLP is released

into the circulation, and the remainder is metabolized to

pyridoxal and 4-pyridoxic acid and also secreted. The major

regulation of PLP concentration in the liver is controlled

by a balance between hydrolysis of PLP by phosphatase and

the protection of PLP against hydrolysis by protein-binding


This type of regulation has also been demonstrated to

exist in other body tissues including erythrocytes (87) and

the brain (95). There is evidence that vitamin B-6 enters

the central nervous system from the plasma by a saturable

transport mechanism (96) and that the choroid plexus serves

as the primary site of entry and is the main source of

phosphorylated vitamin B-6 in the cerebrospinal fluid (97).

The total body store of vitamin B-6 has been reported

to range from 40 to 150 mg pyridoxine equivalents as

estimated by administration of tritium-labeled pyridoxine to

human subjects (98). In this study 2% to 3% of the vitamin

B-6 reservoir was eliminated per day, and the total turnover

rate was 2.2% to 4.4% per day. 4-Pyridoxic acid was the

major urinary vitamin B-6 metabolite excreted, accounting

for 20% to 40% of the isotope eliminated.

In summary, pyridoxal is the form that is transported

across cell membranes into extrahepatic tissues where it is

rephosphorylated and bound to intracellular proteins and

enzymes. Plasma PLP bound to albumin represents a

circulating storage pool since protein-binding effectively

protects it from hydrolysis and the binding capacity with

albumin is large. Metabolism of vitamin B-6 in various

tissues has recently been reviewed (86, 99).

Vitamin B-6 Requirements

The 1980 Recommended Dietary Allowances (RDA) for

vitamin B-6 by the Food and Nutrition Board of the National

Research Council are 2.2 mg/day for men 19 years and older,

2 mg/day for females 15 years and older, 2.6 mg/day for

pregnant women, 2.5 mg/day for lactating women and 0.3

mg/day for infants up to 6 months of age (100). These RDAs

are based on limited information regarding the vitamin B-6

requirements of these groups.

The requirement for vitamin B-6 of humans has been

difficult to establish since clinical manifestations of mild

vitamin B-6 deficiency are nonspecific and not readily

recognizable. Primarily, the requirement for vitamin B-6 has

been estimated by the amount of the vitamin required to

normalize the urinary excretion of tryptophan and methionine

metabolites after tryptophan and methionine loading (100,

101). Since vitamin B-6 functions as a coenzyme in the

metabolism of amino acids, disturbances in amino acid

metabolism have proved to be sensitive indicators of vitamin

B-6 deficiency (102).

Using the method of vitamin B-6 depletion and repletion

with various biochemical measurements, a series of studies

involving small groups of young adult men were conducted by

two research groups: the U.S. Army Medical Research and

Nutrition Laboratory and the University of Wisconsin. The

U.S. Army research group gave a synthetic vitamin B-6

deficient diet containing 0.06 mg of vitamin B-6 to eight

subjects (103) while the Wisconsin group provided a

partially purified diet with 0.16 mg of vitamin B-6 to six

subjects (104, 105, 106, 107). A tryptophan load of 10 mg of

DL-trypophan (U.S. Army) or 2 mg L-tryptophan (Wisconsin)

was administered to challenge the tryptophan metabolic

enzyme systems. Although other biochemical measurements were

determined including blood and urinary concentrations of

various vitamin B-6 vitamerst the vitamin B-6 requirement

was determined from the tryptophan load test since the other

measurements proved to be insufficient to assess the

requirement. In general, blood and urine levels of vitamin

B-6 vitamers and 4-pyridoxic acid decreased during vitamin

B-6 depletion and increased with repletion. The effect of

the level of dietary protein on the vitamin B-6 requirement

was also investigated in these experiments.

In the U.S. Army study, half o4 the subjects were fed a

low protein diet (30 g/day) while the rest were given a

high-protein diet (100 g/day) (103). After three weeks on

the vitamin B-6 deficient diet (0.06 mg/day), the tryptophan

load test was administered, and xanthurenic acid excretion

was measured. Pyridoxine supplementation then was initiated,

and pyridoxine levels were titrated with weekly urinary

xanthurenic acid measurements until the minimal vitamin B-6

requirement was met. The pyridoxine requirement was

estimated to be 1.5 mg/day for the high-protein group and

1.0 mg/day for the low-protein group.

In one experiment, the Wisconsin group found that 0.6

or 0.9 mg/day of supplemental pyridoxine was insufficient to

restore tryptophan metabolite excretion to predepletion

levels in all subjects (104). In a second study, five

subjects were given 54 g/day dietary protein, and six were

fed 150 g/day (107). The high-protein group developed

abnormal metabolism of tryptophan in about one-third the

time that the low-protein group did. In both groups before

depletion, 1.5 mg of pyridoxine were sufficient to maintain

normal excretion levels of tryptophan metabolites; but after

depletion, 1.06 mg of pyridoxine were insufficient to

restore excretion levels to normal. When a similar study was

conducted in which the effect of vitamin B-6 deficiency on

methionine metabolism was investigated, 2.16 mg of

pyridoxine given daily during the repletion period were

barely adequate to normalize methionine metabolism in

subjects consuming 150 g/day of protein.

Taken together, these data indicate that the vitamin

B-6 requirement is between 1.5 and 2.0 mg/day for men

consuming 100 150 g/day of protein and between 1.0 and 1.5

mg/day for men with a dietary protein intake less than 100

g/day. The requirement appears to be greater than 2.0 mg/day

when dietary protein intake exceeds 150 g/day.

The vitamin B-6 requirement of young women was

determined by two research groups, one at Cornell University

using 8 subjects (108) and the other at the University of

Wisconsin with 5 subjects in one experiment (109) and with

10 subjects in a later experiment (110, 111). There were

differences in the experimental methods among the three

studies including dietary protein levels; length of

pre-depletion, depletion and repletion periods; and levels

of dietary vitamin B-6 and supplemental pyridoxine. Despite

these differences, similar conclusions were reached. The

requirement of young women for vitamin B-6 appears to be

between 1.5 and 2.2 mg/day based on the amount of pyridoxine

required to return various biochemical parameters to

pre-depletion levels. Two of the research groups reported

that although aspartate aminotransferase activity in

erythrocytes decreased during vitamin B-6 depletion,

supplementation with pyridoxine up to 2.2 mg/day failed to

restore activity to original levels (108, 110). The

Wisconsin group observed that in vitro stimulation of the

enzyme system by addition of PLP resulted in an increase of

enzyme activity during depletion and a decrease with

repletion although original levels were not reached (110).

The Cornell group found that alanine aminotransferase

activity failed to respond to either vitamin B-6 depletion

or repletion (108) while the Wisconsin group reported a

decrease in activity with depletion and an increase (without

attaining predepletion levels) during repletion (i10).

Stimulation of the enzyme with PLP was variable and

inconclusive. Urinary excretion of 4-pyridoxic acid was

measured in both studies, and 1.0 mg/day of supplemental

pyidoxine was insufficient to restore original levels, 1.5

mg/day was borderline and 2.2 mg/day was excessive (108,

110). Either a 1.0 or 2.2 mg/day intake of pyridoxine was

sufficient to restore urinary tryptophan metabolites to

predepletion levels (109, 111). Results from methionine

loading tests were inconclusive due to variable data (109).

Plasma PLP levels exceeded predepletion levels with 2.0

mg/day of pyridoxine while 0.8 mg only brought plasma PLP to

50% of the original values (111).

These data form the basis of the RDA established for

adult men and women in the United States. The allowance for

vitamin B-6 set by the Canadian Bureau of Nutritional

Sciences is also based on these data but evaluated in an

alternate way. The Dietary Standard for Canada recommends an

intake of 0.02 mg of vitamin B-6 per gram of dietary protein

(112). From the data in the studies reporting both vitamin

B-6 and protein intake, it was calculated that a vitamin

B-6/protein ratio of 0.017 was inadequate while 0.019 was

sufficient to return the majority of biochemical indices to

predepletion values.

The RDA for vitamin B-6 for pregnant women is

extrapolated from the vitamin B-6 requirement of nonpregnant

women. The increase in the RDA during pregnancy from 2 to

2.6 mg is based on the additional vitamin B-6 needed for the

increased protein allowance in pregnancy (100). No

additional vitamin B-6 is recommended to compensate for

fetal demand, increased maternal metabolic requirements and

hormonal induction of maternal vitamin B-6 dependent enzymes

found to occur during pregnancy (32, 33, 113- 15).

Numerous studies have demonstrated that a significant

number of pregnant women who consume a normal diet or

receive vitamin B-6 supplements containing the RDA have low

levels of various biochemical parameters consistent with a

vitamin B-6 deficiency in non-pregnant women (32, 33,

113-114, 116, 117). However, the Committee on Dietary

Allowances of the Food and Nutrition Board has indicated

that although the establishment of the vitamin B-6 RDA

during pregnancy is "beset with uncertainties," there are

insufficient data to justify recommending a higher allowance

which would exceed that provided by the usual diet (100).

The vitamin B-6 RDAs for infants and lactating women

are also established from very limited data. The RDA for

infants is based on the vitamin B-6 content of breast milk

and additional infant foods and the protein intake of the

infants (118). The additional 0.5 mg in the recommended

allowance during lactation above the nonpregnant RDA is

based on information that women consuming vitamin B-6 at

that level seem to provide the needs of the breast-fed

infant (100).

Vitamin B-6 Deficiency

Impairment of growth, the appearance of a pellagra-like

dermatitis (acrodynia), and the development of ataxia are

commonly observed effects of a vitamin B-6 deficiency in

various animal species (70). A hypochromic microcytic anemia

which is responsive to vitamin B-6 administration occurs in

many species (119). Vitamin B-6 deficiency has a profound

effect on the nervous system. In addition to ataxia, these

nervous system abnormalities have been observed in various

species: hyperacousis, hyperirritability, altered mobility

and alertness, abnormal head movement and convulsions (65).

Vitamin B-6 deficiency in the rat also results in muscular

weakness, fatty liver, convulsive seizures, reproductive

impairment, edema, enlarged adrenal glands, and impaired

immune responses (70). Alterations in various biochemical

processes have been observed, including increased excretion

of xanthurenic acid, urea and oxalate, decreased

transaminase activities, reduced synthesis of ribosomal and

messenger ribonucleic acid (RNA) and deoxyribonucleic acid

(DNA) (70).

Clincal manifestations of vitamin B-6 deficiency in

humans are less obvious. A vitamin B-6 deficient diet alone

produces only nonspecific symptoms of deficiency such as

mental depression, irritability and oral lesions. Vitamin

B-6 deficiency has been induced experimentally by the use of

vitamin B-6 antagonists such as 4-deoxypyridoxine. In such

studies, the following symptoms developed in some of the

subjects: oral lesions, dermatitis, peripheral neurophathy,

weight loss, apathy, insomnia, irritability, slight decrease

in immune response, and abnormal electroencephalograms


Dramatic manifestations of a vitamin B-6 deficiency

occurred in human infants fed a proprietary liquid canned

milk formula later found to be vitamin B-6 deficient due to

thermal processing. These symptoms included irritability,

nervousness, convulsive seizures, and abnormal

electroencephalograms which were promptly reversed by

intramuscular injections of 100 mg of pyridoxine (36, 123).

These convulsions are thought to be related to decreased

levels of Y-aminobutyric acid, an inhibitory

neurotransmitter formed from glutamic acid in a vitamin B-6

dependent reaction (124). A report regarding two infants

after several months on a vitamin B-6 deficient diet

indicated that convulsions occurred in one and hypochromic

anemia in the other (125). Both ceased to gain weight.

Urinary excretion of 4-pyridoxic acid ceased while excretion

of vitamin B-6 was reduced to very low levels. Also, the

ability to convert tryptophan to nicotinic acid was


Numerous changes in biochemical processes have been

reported to occur in vitamin B-6 deficienct humans.

Tryptophan metabolism is altered resulting in high urinary

excretion of xanthurenic acid and other metabolites after a

tryptophan challenge (102). This pathway, which requires PLP

as a coenzyme at various points, is illustrated in figure 3.

High dietary intake of protein hastens the onset of vitamin

B-6 deficiency (126). Plasma and blood levels of vitamin B-6

as well ds urinary excretion of vitamin B-6 and 4-pyridoxic

acid are decreased in vitamin B-6 deficiency (102).

Alterations in methionine metabolism have been reported

(127). Incorporation of cystine into hair, which consists of

protein with a high cystine content, is reduced in vitamin

B-6 deficiency which may explain the partial acrodynia

observed in animals (128). Decreased erythrocyte alanine and

aspartate aminotransferase activities and coenzyme

saturation of these enzymes have been observed in vitamin

B-6 deficiency (102). Changes in plasma and urinary levels

of amino acids have also been reported (126).

A number of conditions and diseases may lead to vitamin

B-6 deficiency in humans. Low plasma PLP levels and

decreased activities of aminotransferases have been reported

in alcoholics and patients with liver disease and uremia










hydroxylase ANTHRANILIC


t ransaminase

kynureninase XANTHURENIC





Figure 3. Trytophan to niacin pathway in the human.
Asterisks indicate reactions which require PLP as a

(83, 129, 130). The decreased PLP levels are thought to

result from increased clearance of the coenzyme from the

circulation (130). There is evidence of increased PLP

degradation by the diseased liver (129). Recently,

hydrolysis of PLP in plasma was demonstrated in patients

with liver disease and other conditions with raised alkaline

phosphatase, whether of liver or bone origin (131).

Abnormal responses to tryptophan load tests suggestive

of a vitamin B-6 deficiency have been reported in patients

with a variety of diseases, including Hodgkins' disease,

rheumatoid arthritis, schizophrenia, porphyria,

tuberculosis, aplastic anemia, scleroderma, and various

cancers (132). It has been suggested that liver tryptophan

oxygenase is induced by the increased adrenal secretion of

cortisol brought about by the stress of these diseases


Abnormal tryptophan metabolism as well as low plasma

PLP and erythrocyte aminotransferase activities have also

been observed in pregnant women and oral contraceptive users

(134-136). It has been demonstrated that the abnormal

tryptophan metabolism results from the induction of

tryptophan oxygenase by estrogen, either present in oral

contraceptives or from endogenous secretion in pregnancy

(134, 135). The effects of estrogen on tryptophan metabolism

have been extensively studied and involve both direct

effects on tryptophan metabolism as well as

cortisol-mediated indirect effects (132).

A number of clinical situations exist in which the

requirement for vitamin B-6 is unusually high. Some patients

with the inherited metabolic diseases homocystinuria and

cystathioninuria respond to pharmacological doses of

pyridoxine (137, 138). Patients with a vitamin B-6

responsive sideroblastic anemia require the administration

of 2.5 g or more of pyridoxine daily (70). Certain infants

with convulsive seizures show a high requirement for vitamin

B-6, and dosages between 2 5 mg have relieved convulsive

activity (139).

The effects of vitamin B-6 deficiency on the fetus

during gestation have been extensively studied in animals.

The rat model has been used particularly for experimentation

involving postnatal insults on brain development since the

most rapid increase in the growth rate of the rat brain

occurs approximately 10 days after birth (140). In the human

the most rapid increase in brain growth rate takes place

just before birth.

Early rat studies demonstrated that the progeny of

severely vitamin B-6 deficient pregnant rats exhibited

convulsive seizures, lower birth weights, and lower brain

PLP levels than controls (141). Body weights of litters of

rat dams which received 0, 25 or 50% of the NRC

recommendation for vitamin B-6 were significantly lower than

those of dams receiving 75, 100 or 400% of the NRC

recommended intake (142). Also, neuromotor activity was

delayed in the vitamin B-6 restricted groups. The onset of

crawling was delayed one day for the 25% group versus the

75% group, and standing was delayed four days in the 25%

group compared with the 100% group.

Further studies confirmed these results and indicated

that the impaired neuromotor development of the pups was due

in part to the impaired maternal performance of the dams due

to vitamin B-6 deficiency (143). Pups whose mothers had

received graded levels of vitamin B-6 (0, 25, 50, 75, 100i

or 400% of the NRC recommended intake) were cross-fostered

with a dam isonutritional with the pup's mother or with a

control dam receiving a diet containing 400% of the NRC

recommendation for vitamin B-6. The data indicated that

vitamin B-6 deprivation during gestation operated in two

ways: directly, by impairing the development of the fetus;

and indirectly, by adversely affecting the ability of the

dam to function as a mother which was not easily reversed by

an adequate post-parturition diet.

A similar experiment indicated that the effects of

vitamin B-6 deprivation during lactation generally were not

as severe as those seen during gestation (144). Rats which

received 400% of the NRC recommended intake for vitamin B-6

during gestation were fed 0, 25, 50, 75, 100, or 400% of the

NRC recommendation during lactation. Pups suckling dams

which received 100% or less of the NRC recommended intake

during lactation were later in the onset of the more

advanced neuromotor skills than those whose mothers received

the 400% diet. These data indicated that the prenatal

vitamin B-6 requirement of rats is greater than the

postnatal requirement and that the current NRC recommended

intake of vitamin B-6 for rats may be insufficient for the

optimum growth and development of the progeny.

A 30 to 50% reduction in cerebral sphingolipid level

has been reported in vitamin B-6 deficient suckling rats as

well as severe deviation of cerebral free amino acids from

the normal (145). Sphingolipids are essential for the proper

development of the nervous system, particularly for

myelination. It has been postulated that sphingosine

synthesis is reduced in vitamin B-6 deficiency due to

decreased activity of the vitamin B-6 dependent synthase

enzyme which forms 3-dehydrosphinganine from palmitoyl CoA

and serine.

Vitamin B-6 deficient suckling rats exhibited reduced

recoverable myelin in the brain which was not attributable

to decreased brain weight alone (146). Myelin recovery,

total myelin lipids, and cerebral vitamin B-6 content were

inversely related to the duration of the vitamin B-6

deprivation. The polyunsaturated fatty acid content of the

myelin phospholipids was markedly reduced in severely

vitamin B-6 deficient pups. However, the relationship

between the synthesis of long-chain polyunsaturated fatty

acids and vitamin B-6 is not known. The activity of

3-dehydrosphinganine synthase in the vitamin B-6 deficient

pups was only 62% of the control group. However, enzyme

activity increased to normal levels when PLP was added in

vitro, demonstrating that the reduced enzyme activity was

due to vitamin B-6 deficiency and not to lack of apoenzyme.

Twelve-day old rat pups whose dams were fed low levels

of vitamin B-6 throughout growth, gestation and lactation

exhibited reduced brain levels of vitamin B-6 and

cerebrosides (147). Total lipid, phospholipid, cholesterol

and proteolipid content were similar in all treatment

groups. Brain lipid profiles of progeny of vitamin B-6

deficient rats before and after vitamin B-6 supplementation

were studied (148). Cerebroside and ganglioside content were

significantly reduced in pups from unsupplemented vitamin

B-6 deficient dams and from dams supplemented beginning ten

days postpartum. Supplementation of vitamin B-6 deficient

dams five days postpartum normalized levels of these brain

lipids in the offspring.

Cytoarchitectural alterations in the brains of rat pups

of vitamin B-6 deficient dams have been observed, including

reduced neocortex and cerebellum areas and adversely

affected Purkinje cell differentiation (149). Although

maternal vitamin B-6 deficiency during the period of rapid

myelination (10 to 20 days postnatal in the rat) did not

result in decreased myelination in the brains of progeny

(i50), decreased myelination in the spinal cords of progeny

of vitamin B-6 deficient dams has been observed (151).

Administration of the vitamin B-6 antagonist

4-deoxypyridoxine to rat dams deprived of vitamin B-6 during

gestation inhibited replication and function of lymphocytes

involved in cell-mediated immunity of the progeny (152).

Passive antibody transfer during lactation from dams to rat

pups as measured by immunoglobulin G concentrations in

maternal serum, pup serum, and milk was not affected by

maternal vitamin B-6 deficiency (153). However, suckling

rats of vitamin B-6 deficient dams and weanling rats

consuming a vitamin B-6 deficient diet had fewer spleen

cells, splenic antibody-forming cells and lower levels of

circulating antibodies than non-deficient groups. This

reduction resulted partly from lower food intake but was

intensified by vitamin B-6 deprivation.

Methods of Vitamin B-6 Status Assessment
and Vitamin B-6 Analysis

Methods to assess vitamin B-6 status rely mainly upon

dietary intake and biochemical data since clinical signs of

vitamin B-6 deficiency are nonspecific and are rarely seen

in a free-living population. Biochemical methods which have

been used to evaluate vitamin B-6 status include blood and

urinary levels of the vitamin B-6 vitamers or metabolites,

activities of vitamin B-6 dependent enzymes in blood, and

levels of urinary metabolic products (102).

Estimation of dietary intake of nutrients, including

vitamin B-6, has been limited for practical reasons to two

methods: 24-hour dietary recalls or 3-day diet records. The

24-dietary recall method has been criticized for a variety

of reasons (154, 155). However, when the technical

limitations are recognized in interpreting data from

24-dietary recalls it is generally agreed that 24-hour

recalls provide estimates of group intakes that are

comparable to results obtained with more cumbersome

techniques (156). It has been pointed out that while the

mean and median intake values for the group obtained by

24-hour dietary recall method have validity, it is incorrect

to assign percentile positions or ranks to individuals in

the group and then generalize about the individual (157). A

statistical examination of sources of variance in 24-hour

dietary recall data has been performed (156). Potential

sources of error in the recall method include the recall

ability of the subject and the accuracy and completeness of

food composition data.

Most reports which have included vitamin B-6 dietary

intake information have used the 24-dietary recall (117,

136, 158). Roepke and Kirksey (159) compared the results

obtained from 24-dietary recalls and 3-day diet records and

found a correlation of 0.78 for the vitamin B-6 intake

estimated by the two methods. The 3-day diet record provided

a slightly larger estimate of intake than the 24-dietary


Vitamin B-6 status has been assessed with direct

methods by measuring vitamin B-6 vitamers in blood, urine

and other tissues. Indirect methods which have been used

include measurement of tryptophan or methionine metabolite

excretion following a loading dose of the amino acid and

determination of activities of vitamin B-6 dependent enzymes

before and after adding PLP. The tryptophan load test has

been more frequently used than the methionine loading

procedure. After a 2 or 5 gram oral loading dose of

L-tryptophan, urinary xanthurenic acid, 3-hydroxykynurenine,

and kynurenine are measured by spectrophotometric,

colorometric or fluorometric methods following separation by

column chromatography (111). Although this method provides a

measure of the functional adequacy of coenzyme levels, many

factors other than vitamin B-6 status can influence urinary

excretion of these metabolites including protein intake,

lean body mass, exercise, individual variations, the size of

the amino acid loading dose, and estrogen levels (102).

Another disadvantage is that this procedure requires a

24-hour urine collection which is often not practical.

Consequently, this method is not used currently although

many of the vitamin B-6 requirement studies utilized this

procedure (33). Because of the marked derangement of

tryptophan metabolism caused by estrogen through the

induction of tryptophan oxygenase, the tryptophan load test

is not useful in assessing vitamin B-6 status in pregnant

women or oral contraceptive users.

Transaminase activities in blood also provide a

functional measurement of the state of vitamin B-6 reserves.

Transaminase activities have been measured in erythrocytes,

leukocytes and plasma. Since plasma transaminase activities

are much lower than in erythrocytes and show a wide range in

normal individuals, they are not considered useful in

assessing vitamin B-6 status. The two transaminases measured

in erythrocytes are alanine aminotransferase (AlaAT, E.C. and aspartate aminotransferase, (AspAT, E.C., also referred to as glutamic pyruvate transaminase

(GPT) and glutamic oxaloacetate transaminase (GOT),


Several reports have indicated that measurement of

these enzyme activities per se did not give consistent

results but that the degree of stimulation by PLP added in

vitro was a better indicator of vitamin B-6 status

(160-162). More recently, Lumeng et al. (163) demonstrated

that in general erythrocyte AspAt and AlaAT activities were

more sensitive indicators of vitamin B-6 status than

stimulation values. AspAT activity has been reported to be

higher than AlaAT in all tissues measured in the rat and is

therefore more easily determined (164). There are

conflicting reports as to which enzyme provides the better

measurement. Cinnamon and Beaton (160) found that

erythrocyte AlaAT stimulation by PLP was more sensitive to

dietary vitamin B-6 depletion and repletion than stimulation

of AspAT. However, Donald et al. (108) found the converse to

be true. In a recent report of the vitamin B-6 requirement

of oral contraceptive users assessed by erythrocyte AspAt

and AlaAT activities and stimulation with PLP, Bosse and

Donald (165) observed random fluctuations in AspAT activity

during vitamin B-6 depletion and repletion. Although

erythrocyte AlaAT activity was sensitive to vitamin B-6

deficiency, it was slow to respond to pyridoxine


A number of factors may contribute to the decreased

sensitivity of the aminotransferase activity measurements

and stimulation values, including variations in the assay

procedures used (166). Normal healthy individuals appear to

have a wide range of activity and stimulation values (102).

Stimulation of aminotransferase activities may be substrate

concentration dependent (167). Sample handling may be a

factor since intracellular changes in the degree of

saturation of the apoenzyme after cell death have been

observed (168). Since PLP is tightly bound to both enzymes,

saturation decreases only in relatively severe vitamin B-6

deficiency. Transfer of PLP to AspAT and AlaAT from

low-affinity binding proteins such as the vitamin B-6

dependent enzymes serine dehydratase and tyrosine

aminotransferase has been shown to occur in rat liver (163).

Some aminotransferases appear to be induced by hormones


Consequently, erythrocyte AlaAT and AspAT activity and

stimulation measurements have not been significantly

correlated with other parameters of vitamin B-6 status such

as plasma PLP levels in numerous studies, particularly in

pregnant subjects (113, 117, 170, 171). Reasonable

correlations have been obtained when vitamin B-6 deficiency

is relatively severe or pyridoxine supplementation is given

(113, 163).

The AspAT and AlaAT activities have been assayed

colorimetrically (166, 172), but these procedures have been

criticized for their lack c- specificity (173, 174). The

recommended method for measuring serum AspAT and AlaAT

activities is a coupled enzyme spectrophotometric procedure

which has been adapted to erythrocytes (175). In this method

the oxaloacetate or pyruvate formed from the transanimase

reaction is reduced by NADH2 in a malic or lactic

dehydrogenase catalyzed reaction, and the reaction rate is

monitored by a decrease in absorbance at 340nm as NADH2 is

oxidized. Recently, this procedure for measuring AspAT and

AlaAT activities has been optimized (176) and adapted for

use in automated systems (177).

Measurements of the major urinary excretory metabolite

of vitamin B-6, 4-pyridoxic acid, and to a lesser extent,

free vitamin B-6, have been used to assess vitamin B-6

status. It has been assumed that excretion of 4-pyridoxic

acid reflects vitamin B-6 intake since urinary levels

decrease during vitamin B-6 depletion and increase again

during repletion (105, 106, 108, 110). However, the

influence of dietary intakes of graded levels of vitamin B-6

on urinary excretion has not been investigated.

The use of urinary excretion of 4-pyridoxic acid

measurements in assessing vitamin B-6 status has been

criticized because 4-pyridoxic acid can be produced by

nonspecific enzymes, and therefore its excretion would not

necessarily be related to vitamin B-6 status (32). Recently,

in a short-term metabolic study, 4-pyridoxic acid excretion

did not correlate with dietary vitamin B-6 intake,

erythrocyte AspAT activity and stimulation by PLP added in

vitro, or plasma PLP levels (171).

Traditionally, 4-pyridoxic acid has been measured

fluorometrically after extensive sample preparation by

ion-exchange chromatography to remove interfering compounds

present in the urine (178, 179). Recently, a simple, rapid

and sensitive high performance liquid chromatographic method

has been developed (180).

Plasma, erythrocyte, and whole blood levels of vitamin

B-6 also decrease rapidly during vitamin B-6 depletion and

increase after repletion (105, 106, 108), but these

measurements have not been successfully used to determine

vitamin B-6 status. Vitamin B-6 vitamers traditionally have

been measured by microbiological methods using mainly

Saccharomyces uvarum (181), although procedures employing

Koeckera brevis have been described (182-184). There have

been reports that S. uvarum has a greater growth response to

pyridoxine than to pyridoxal or pyridoxamine (181, 185-187)

and that K. brevis exhibits an even greater disparity (186).

For individual vitamin B-6 vitamer measurement, column

chromatography is used for separation of the vitamers prior

to microbiological determination (185). A recent preliminary

report regarding the use of high performance liquid

chromatography in measuring vitamin B-6 vitamers in human

plasma seems promising (188).

Total vitamin B-6 levels tend to be a poorer indicator

of status than plasma PLP levels since individual levels of

vitamin B-6 vitamers other than PLP fluctuate considerably

(171). Plasma PLP levels are currently considered to be the

best measurement of vitamin B-6 status in humans. PLP is the

functional form of the vitamin and represents a rapidly

mobilizable storage pool of vitamin B-6 as well as the major

transport form (163). Animal studies indicate that plasma

PLP is derived solely from the liver and that plasma PLP

levels correlate well with vitamin B-6 intake and PLP tissue

levels, particularly with PLP content of skeletal muscle

which is the largest repository of vitamin B-6 in the body

(189). In humans plasma PLP remains relatively constant over

time (81, 82, 110). When vitamin B-6 dietary intake is

altered, plasma PLP levels are changed in accordance with

intake and reach new steady-state levels in 3 to 4 weeks

(110). The menstrual cycle and chronic use of oral

contraceptives does not affect plasma PLP levels in most

females (110, 190).

Various enzymatic assays have been used to measure

plasma PLP (191-198), but the tyrosine apodecarboxylase

procedure is recommended as the method of choice (199). In

this procedure the amount of 14C02 evolved from the

PLP-dependent decarboxylation of uniformly labeled

14C-tyrosine in the presence of PLP and tyrosine

apodecarboxylase is determined by liquid scintillation

spectrometry (191, 192).

Vitamin B-6 Status and Requirement During Pregnancy

Numerous studies have demonstrated that a significant

number of pregnant women exhibit a relatively poor vitamin

B-6 status as indicated by various biochemical tests. When

the vitamin B-6 status of 458 pregnant women was assessed by

the in vitro stimulation of erythrocyte AspAT activity by

PLP, 40 to 60% were in suboptimal status as compared to a

control group of 300 male and female blood donors (ii4). It

was suggested that vitamin B-6 supplementation was required

by 50% of the pregnant women studied in order to maintain

normal coenzyme saturation of the enzyme. When measuring

whole blood PLP levels in 10 pregnant women, Shane and

Contractor (113) reported significantly lower levels

(5.1+1.3 ng/ml, mean+SD) than those of 12 nonpregnant women

(9.6+_ 1.7 ng/ml) and 9 oral contraceptive users

(7.6+1.1 ng/ml).

Sixty-eight percent of 127 low-income pregnant

adolescent and adult women exhibited poor vitamin B-6 status

during the first trimester as measured by PLP stimulation of

erythrocyte AlaAT activity by in vitro addition of PLP

(200). When a subsample was evaluated again at 30 weeks

gestation after participation in the Special Supplemental

Food Program for Women, Infants and Children (WIC), 63% were

in suboptimal vitamin B-6 status. Kaminetzky et al. (201)

reported low blood PLP levels in a significant number of 246

pregnant teenagers. All the mothers of low birth-weight

infants in that study exhibited low PLP levels.

Contractor and Shane (32) investigated the effect of a

50 mg oral dose of pyridoxine on the blood PLP levels of the

mother and fetus as well as nonpregnant women. Initially,

the PLP levels of the pregnant women were significantly

lower' than the nonpregnant controls, indicating a relative

vitamin B-6 deficiency. Also, the cord blood PLP levels were

significantly higher than the maternal blood levels which

suggested active transport of vitamin B-6 across the

placenta. Significant differences were also found between

the low blood PLP levels of the pregnant women and the

higher PLP levels of the nonpregnant women at various times

after oral loading with vitamin B-6. These observations

could be explained by active transport of the vitamin to the

fetus and increased uptake by maternal tissues during

pregnancy. These factors are believed to be responsible for

the relative vitamin B-6 deficiency observed during


Brin (33) obtained similar results when measuring

plasma PLP and erythrocyte AspAT and AlaAT activities in

maternal and cord blood. He pointed out that the maternal

deficiency could easily result from fetal sequestration of

the vitamin since the fetal compartment, which accounts for

10 to 20% of the pregnant woman's physiological mass, can

concentrate vitamin B-6 two to threefold. This would be true

particularly if the mother has a low dietary vitamin B-6

intake or has marginal vitamin B-6 status at the onset of


The effects of vitamin B-6 supplementation during

pregnancy on maternal and fetal plasma PLP levels at term

were studied by Cleary et al. (116). Compared with the mean

plasma PLP level of 58 nonpregnant women serving as the

control group, 4 of 10 pregnant women receiving 10 mg/day of

vitamin B-6 and 10 of 13 receiving 2 to 2.5 mg/day had low

PLP levels. The cord plasma PLP levels of the pregnant women

with adequate plasma PLP levels were compared with the cord

levels of mothers who had low plasma PLP levels. The low PLP

group had only 53% of the PLP cord concentration of the

normal group, indicating that the vitamin B-6 status of the

mother significantly affects the plasma PLP levels of the

fetus. Hamfelt and Tuvemo (202), measuring plasma PLP levels

and in vitro stimulation of erythrocyte AlaAT by PLP, also

reported that a minimum of 10 mg/day of vitamin B-6 may be

needed to raise maternal and fetal vitamin B-6 levels to

normal non-pregnant levels.

By supplementing 26 pregnant women with 2.5, 4 or 10 mg

of pyridoxine daily, Lumeng et al. (117) demonstrated that

plasma PLP levels in all supplementation groups reached a

maximum at 13 to 18 weeks gestation and fell to a minimum at

term. All subjects taking 2.5 mg/day of vitamin B-6 and two

of the 6 subjects taking 4 mg/day developed low plasma PLP

levels during the third trimester, and all had low PLP

levels at term. In this study, low plasma PLP levels were

defined as <4.7 ng/ml, the value 2 standard deviations below

the mean of a nonpregnant control group. One of the 10 women

receiving 10 mg of vitamin B-6 daily had plasma PLP in the

deficiency range. There was no significant difference in the

effect of supplementation between the 2.5 and 4 mg on plasma

PLP levels throughout pregnancy. Stimulation of erythrocyte

AlaAT and AspAT activity by in vitro addition of PLP was

also measured, but these values changed inconsistently with

time and did not correlate well with plasma PLP

measurements. These results indicate that more than 4 mg of

vitamin B-6 daily is required by pregnant women in addition

to usual dietary sources to prevent the low levels of plasma

PLP indicative of vitamin B-6 deficiency at term.

The relationships between vitamin B-6 intake, vitamin

B-6 levels in maternal serum, cord serum and milk was

recently investigated in 106 pregnant and lactating women by

Roepke and Kirksey (159). The vitamin B-6 intake from

unsupplemented diets was less than two-thirds of the RDA

during pregnancy. Subjects consuming 2.5 mg/day or less had

significantly lower serum vitamin B-6 levels at delivery

than those consuming more than 2.5 mg. At delivery the

maternal serum vitamin B-6 levels were lower in mothers

whose infants had Apgar scores less than seven at one minute

after birth than in those whose infants scored seven or

better. In assessing the newborn, Apgar scores are

determined by numerical values given for heart rate,

respiratory effort, muscle tone, reflex irritability, and

color for a maximum score of 10 (203, 204). Maternal serum

vitamin B-6 levels at 5 months gestation significantly

correlated with vitamin B-6 levels in the cord serum at

delivery and in milk 14 days postpartum. Since this stage of

gestation precedes the period of rapid growth in the fetal

central nervous system, these researchers observed that 5

months gestation is a critical time for assessing maternal

vitamin B-6 status.

These data also demonstrate the effect of the vitamin

B-6 nutritional status and intake of the mother during

pregnancy and lactation on the vitamin B-6 content of the

mother's milk. It has been clearly shown that the vitamin

B-6 content in human milk responds rapidly to changes in

vitamin B-6 intake since there are no apparent regulatory

mechanisms to maintain the vitamin B-6 concentration within

definite limits (205, 206).

The condition of the infant at birth has also been

related to the vitamin B-6 status of the mother during

pregnancy by Schuster et al. (200). Infants whose mothers

exhibited poor in vitamin B-6 status at around 15 weeks

gestation had significantly lower Apgar scores at one minute

after birth than those whose mothers exhibited adequate

vitamin B-6 status.

Long-term oral contraceptive use has been implicated in

compromising the vitamin B-6 status of women during

pregnancy and lactation. Roepke and Kirksey (207) reported

that vitamin B-6 levels in maternal serum at 5 months

gestation and delivery and in cord serum and milk were lower

in the long-term >30 months) oral contraceptive users than

in short-term users or non-users. At 5 months gestation the

mean maternal serum level of long-term oral contraceptive

users was only 49% of the mean for non-users and 59% of the

mean for short-term users. These data suggest that long-term

use of oral contraceptives prior to conception may reduce

maternal vitamin B-6 reserves.

Serum PLP levels of premature infants have been found

to be extremely low (mean = 0.82 ng/ml + 0.97 SD) at birth

(208). This observation suggests that the most rapid fetal

uptake of vitamin B-6 may take place during the last weeks

of pregnancy, providing one reason for the dramatic decrease

in maternal plasma PLP levels during the last trimester,

particularly at delivery. Another factor which may

contribute to this phenomenon is the hemodilution which

peaks at this stage of pregnancy (209).

Vitamin B-6 nutritional status in pregnancy may also be

related to several poorly understood disorders of pregnancy,

including morning sickness (nausea and vomiting in early

pregnancy), gestational diabetes, and hypertensive disorders

of pregnancy such as pre-eclampsia. The majority of pregnant

women experience nausea and/or vomiting during the first

trimester of pregnancy (210). The interest in a possible

relationship between vitamin B-6 and morning sickness dates

back 40 years when altered tryptophan metabolism was

observed in pregnant women, particularly those with morning

sickness. This led to the inclusion of pyridoxine in

antinausea prescription drugs such as Bendectin used to

treat the symptoms of morning sickness. Studies dating back

20 to 40 years ago reported conflicting results about the

effect of vitamin B-6 therapy for morning sickness


In 1974 Reinken and Gant (216) determined the serum PLP

levels before and after treatment with 100 mg of pyridoxine

daily for 7 days in 24 pregnant women who were experiencing

vomiting. Pregnant women who were not experiencing morning

sickness had higher levels of serum PLP than the

experimental subjects. The serum PLP levels of the

experimental subjects in the first trimester were comparable

to levels found in healthy pregnant women in the third

trimester of pregnancy. After treatment, the serum PLP

levels in the subjects increased dramatically. The

researchers claim that there was also an alleviation of the

clinical symptoms of morning sickness, but no data were

presented to substantiate this. In fact, it is hardly

surprising that women who experience daily vomiting in

contrast to the milder symptoms of morning sickness would be

poorly nourished with respect to vitamin B-6 and other

nutrients due simply to the significant loss of nutrients

through vomiting. Unfortunately, this poorly-designed study

has not clarified any possible relationship between vitamin

B-6 and morning sickness.

A double-blind study was conducted by Wheatley (217) in

1977 to compare the effects of the antinausea drug Debendox

(the British equivalent of Bendectin) with 10 mg of extra

pyridoxine with a placebo containing 10 mg of pyridoxine.

The Debendox preparation was more effective in decreasing

the frequency of nausea and the severity of nausea and

retching but was no more effective than pyridoxine in

decreasing the severity of vomiting. The significance of

these findings is difficult to assess. These data do

indicate that Debendox is not effective in treating the most

severe symptoms of morning sickness.

Gestational diabetes mellitus is a condition associated

with high perinatal mortality rates which becomes manifest

during the stress of late gestation. A possible relationship

between diabetes and vitamin B-6 was first postulated when a

high-tryptophan low-vitamin B-6 diet caused diabetes to

develop in rats (218, 219). It was demonstrated that a

complex between insulin and the tryptophan metabolite

xanthurenic acid was formed which reduced insulin activity.

The proposed mechanism is a coupling of the xanthurenic acid

at the histadine-zinc group to insulin (220). Xanthurenic

acid binds to both albumin and insulin in the serum, and the

amount of xanthurenic acid bound to serum protein is greater

in normal serum than in diabetic serum (221). The glucose

tolerance curves of 13 pregnant women with gestational

diabetes in late pregnancy improved after treatment with 100

mg/day of vitamin B-6 for two weeks (222). Insulin levels

remained the same or decreased at all time-points on the

two-hour curved suggesting that the treatment increased the

biological activity of the plasma insulin. However, no

measurements to determine vitamin B-6 status in these

patients were made.

Pre-eclampsia is characterized by hypertension,

proteinuria and edema which appear after the 20th week of

gestation. It usually occurs in the very young or old

primigravidas and can be fatal. Levels of vitamin B-6

vitamers in pre-eclamptic placentae have been reported to be

only 40% that of normal placentae (223). Pre-eclamptic

newborn had less than 50% the cord blood PLP concentration

of normal infants (224). Subjects in a survey of 246

pregnant teenagers who developed pre-eclampsia all exhibited

low blood PLP levels (201).

Low fetal plasma PLP levels may result from decreased

fetal synthesis of PLP or from decreased placental uptake of

maternal PLP. Low activities of two enzymes involved in

vitamin B-6 metabolism in pre-eclamptic placentae have been

reported. Pyridoxal kinase and pyridoxamine (pyridoxine)

phosphate oxidase activities were reduced in pre-eclamptic

placentae while phosphatase activity was not (223). The

placental uptake of other nutrients which involve active

transport such as amino acids has been shown to be decreased

in pre-eclampsia (225).

Reports regarding the effect of vitamin B-6

supplementation on the incidence of pre-eclampsia have been


contradictory. Hillman et al. (226) found no significant

difference in the incidence of pre-eclampsia when vitamin

B-6 alone was given to a low socioeconomic group. However,

the incidence of pre-eclampsia was reduced by 50% when

vitamin B-6 was added to a general vitamin regime in

middle-class subjects (227).



The subjects in this study were clients attending

Maternal and Infant Care (MIC) clinics located in Alachua,

Marion and Putnam counties in north central Florida between

June 1981 and April 1983. The MIC clinics are state-funded

facilities which provide prenatal and postpartum care for

eligible low-income women and their infants. Two hundred and

forty women volunteered to participate in the study during

their first prenatal clinic visit. The group was 48% black

and 52% white. Subsequently, 47 subjects stopped attending

the clinic, 3 were not pregnant, 4 miscarried, 4 delivered

prematurely, and 65 discontinued taking the vitamin B-6

supplements. Blood samples were obtained from 71 subjects at

the 30-week gestation clinic appointment, and 40 maternal

and cord blood sample sets were obtained at delivery.

A nonpregnant group of healthy women was used to

provide normal nonpregnant values of plasma PLP levels and

erythrocyte AspAT activity and stimulation by exogenous PLP.

Twenty-seven well-nourished students and staff members at

the University of Florida volunteered to provide one 10-ml

fasting blood sample. These women, ranging in age from 19 to

34 years, did not take oral contraceptives or vitamin B-6


Experimental Design

Subjects were randomly assigned a vitamin B-6

supplement containing 0, 2.6, 5, 7.5, 10, 12.5, 15 or 20 mg

of pyridoxine-HC1 at their first prenatal clinic visit.

Since this was a double-blind study, the supplements were

identified by a letter code in order that neither subjects

nor researchers knew what level was being administered. The

key to the code was not released until the study was

finished and all samples were analysed. The vitamin B-6

supplements were provided by the Hofmann-LaRoche Co.

(Nutley, New Jersey). The supplements were in tablet form

and contained lactose, microcrystalline cellulose, corn

starch, magnesium stearate, and pyridoxine-HCl.

Maternal health and vitamin B-6 status were assessed at

three stages of pregnancy: first prenatal clinic visit (<22

weeks gestation), 30 weeks gestation, and at term. Vitamin

B-6 status and condition of the infant at birth were also

determined. Dietary vitamin B-6 intake of each subject was

estimated from a 24-hour dietary recall.

Research Protocol

First Prenatal Clinic Visit

Eligibility of the subjects to participate in the study

was determined after completion of their medical history and

physical examination. Criteria for eligibility were i) good

health at first visit, 2) less than 22 weeks pregnant, 3) 17

years or older, 4) not taking vitamin B-6 supplements or

medications such as Bendectin which contain vitamin B-6, 5)

and no long-term history of oral contraceptive use (more

than one year within 6 months of current pregnancy).

The purpose and protocol of the study were explained to

each eligible client, and each was asked if she wished to

volunteer to participate in the study. A study description

and instruction form (form i in Appendix) was given to each

volunteer along with a bottle containing 200 supplements.

Each subject was also verbally instructed to take one tablet

daily and to bring the bottle with any remaining supplements

to the 30-week gestation clinic appointment. Informed

consent forms (form 2 in Appendix) were read and signed by

the subjects, and a copy was given to each one.

Subjects were interviewed regarding morning sickness

(nausea and vomiting) experiences and the use of alcohol,

tobacco and other drugs. Other personal and medical data

were taken from the subject's medical record, including

hematocrit, blood pressure, height, weight, pre-pregnancy

weight, age, race, gestational age, and parity. A 10-ml

blood sample was obtained from each subject by the clinic


Thirty-week Clinic Visit

At 30 weeks gestation, each client attending the

prenatal clinic is given a glucose tolerance test to screen

for diabetes mellitus. At this appointment a 10-ml blood

sample was also obtained from subjects in this study. The

subject was interviewed, and a new bottle of vitamin B-6

supplements was provided. Each subject was questioned

regarding her degree of compliance with supplementation. In

addition, the number of pills remaining in the old bottle of

supplements returned by the subject was counted. A 24-hour

dietary history was obtained. Medical record data were

recorded, including weight, hematocrit, blood pressure, and

glucose tolerance.


The majority of the subjects delivered at Shands

Teaching Hospital on the University of Florida campus, and

some of the low-risk clients delivered at Putnam County

Hospital or Munroe Regional Medical Center (Marion County)

through the midwifery program. A 10-ml blood sample was

obtained from each subject during delivery. A cord blood

sample was collected after ligation of the cord. Data

regarding maternal health status were obtained from the

medical record, including hematocrit, blood pressure, and

information regarding complications during pregnancy and

delivery. Placental weight was also recorded. Information

obtained regarding the condition of the infant at birth

included length, weight, Apgar scores at I and 5 minutes

after birth, and any abnormalities or health problems.

Sample Collection

All blood samples obtained from the subjects were

collected by venipuncture in sodium herparinized Vacutainer

tubes. Cord blood was also collected in heparinized tubes.

The samples obtained at the initial and 30-week visits were

immediately centrifuged at approximately 1000 x g for 15

minutes, and the plasma was removed and saved. The

erythrocytes were washed in an equal amount of 0.85% saline

solution and again centrifuged for 15 minutes. The

supernatant was removed, and the erythrocytes transferred to

storage vials. Sample handling was carried out in subdued

light to minimize destruction of vitamin B-6. All samples

were stored on ice in a covered container while transported

to the university laboratory and then stored at -300C until

analyzed. Maternal and cord blood samples obtained at

delivery were also centrifuged for i: minutes, and the

plasma was transferred to storage vials and immediately


Biochemical Analyses

Aspartate Aminotransferase Activity

Erythrocyte AspAT activity and in vitro stimulation by

exogenous PLP were assayed using a continuous flow procedure

developed by Skala et al. (177) with several modifications.

This method is based oh the following reaction sequence:

Aspartate + 2-Oxoglutarate Oxaloacetate + Glutamate

Oxaloacetate + NADH ---- Malate + NAD

where MDH is malic dehydrogenase and NADH and NAD are the

reduced and oxidized forms of nicotinamide adenine

dinucleotide, respectively.

AspAT activity is proportional to the rate of oxidation

of NADH which can be measured by the rate of decrease in

absorbance at 340 nm or fluorescence at 470 nm. Lactic

dehydrogenase is added to reduce endogenous pyruvate, a

source of competing side reactions.

An AutoAnalyzer I (Technicon Instruments, Tarrytown,

NY) was used in this study. This system measures the

endpoint of the indicator reaction. It was therefore

calibrated with a series of control red cell hemolysates,

the AspAT activity of which were measured by a kinetic

method performed manually which will be described later.

These hemolysates served as standards and were included with

every 30 samples analyzed by the AutoAnalyzer I system.

The reagents used for the continuous flow method and

their concentrations are listed in table I and were obtained

from Sigma Chemical Co. All reagents except PLP were

prepared with the Tris buffer solution and brought to pH

7.8. Triton X-100 was added to the aspartic acid and

NADH/MDH solutions to give a 0.1% final concentration. Tris

buffer was added to erythrocytes to provide a 1/15 dilution.

Erythrocytes were stored at -300C for i 10 months prior to


The integration of the reactants into the system is

shown in figure 4. The hemolysate samples were incubated

with aspartic acid at 300C. PLP was added at this point to

measure in vitro stimulation of AspAT activity by exogenous

PLP. Water was substituted for the PLP to measure basal

AspAT activity. The reaction was initiated by

2-oxoglutarate, as recommended by the International

Federation of Clinical Chemistry (IFCC) (228). The resulting

oxaloacetate was dialyzed into the recipient stream. The

NADH/MDH solution was introduced and the indicator reaction

took place on the recipient side of the dialyzer. After

incubation at 300C for approximately 5 minutes, the NADH

concentration was measured fluorometrically (Aminco

FluoroMonitor) and recorded using a strip chart recorder.

AspAT activity is linearly proportional to the decrease in

fluorescence as NADH is oxidized to NAD, which unlike NADH

does not fluoresce at 470 nm when excited at 340 nm. A 0

Table i. Reagent concentrations for continuous flow analysis
of AspAT activity.


Reagent Working Fina l
Reagent Reaction

Tris hydroxymethyll
aminomethane) 104 82 (m M)

L-Aspartate 505 200 (mM)

NADH 0.12 0.12 (mM)

MDH 610 610 (U/ )

2-Oxoglutarate 80 12 (mM)

PLP 0.66 0.10 (mM)

Volume fraction
of sample 0.25 (0.025-0.04)

pH 7.8 7.8

" E
* "

- z <
0 3
U .
o I E

4 ( zI




c4 (ud D

4c .


d -









0 *

4- >



_ 0



fluorescence baseline was established by substituting water

for the NADH entering the system.

The AspAT activities of the hemolysates used as

standards to calibrate the continuous flow method were

determined manually immediately prior to each series of

determinations by the AutoAnalyzer I. Due to insufficient

supply of control erythrocytes, different ones were used

each time. However, a sufficient amount of one sample

allowed for its measurement each time to study the stability

of erythrocyte AspAT activity with time under the storage

conditions of the study.

The assay conditions used were those optimized and

recommended by Bergmeyer et al. (176) for determination of

serum AspAT activity. The reagents used and their final

concentrations are listed in table 2. The hemolysates were

prepared by adding deionized distilled water to thawed

erythrocytes to yield a 1/45 final dilution and kept on ice.

All samples were analyzed in triplicate. The reagent mixture

(2.0 ml) was incubated with the hemolysate (0.2 ml) for 10

minutes at 3000. Then the reaction was started by adding

2-oxoglutarate (0.2 ml) pre-warmed to 300C. The reaction was

followed spectrophotometrically (Model 250, Gilford

Instruments) at 340 nm for 10 minutes. The initial reagent

mixture consisted of 1.9 ml Tris buffer, 0.05 ml NADH/MDH

solution and 0.05 ml LDH enzyme (from rabbit muscle, Sigma

Chemical Co., 340-10) when measuring basal AspAT activity.

Stimulation of AspAT activity was accomplished by adding 0.1

Table 2. Assay conditions and final reagent concentrations
for spectrophotometric determination of AspAT activity by
coupled-enzyme kinetic method.



Tris (hydroxymethyl-







30 oC


80 mM

240 mM

12 mM

0.18 mM

420 U/1

600 U/I

0.10 mM

Volume fraction of sample

0.083 (1:12)

ml of PLP solution and decreasing the buffer volume to

1.8 ml.

AspAT activity is expressed in international units per

liter of hemolysate and is calculated using the following


dA/min x 2.4 x 1000 x TCF
Activity(U/L) =
6.22 x 0.20

where dA/min is the change in absorbance per minute, 2.4 is

the toal reaction volume (ml), TCF is the temperature

correction factor which is 1.00 at 300 C, 6.22 is the

millimolar absorptivity of NADH at 340 nm, and 0.20 is the

sample volume (ml).

The linear range of the assay was determined by

analyzing serial dilutions of a control hemolysate. Analysis

of another hemolysate stored at -300 C indicated little

change (<5%) in AspAT activity over a period of 10 months.

AspAT activity was 723.5 U/L with 38% stimulation by added

PLP in the first month, 737.9 U/L with 40% stimulation in

the fifth month, and 710.2 U/L with 35% stimulation in the

tenth month. The mean correlation coefficient (r) of the

plots of the standard hemolysates assayed manually versus

the continuous flow method was 0.953 as determined by linear

regression analysis. This highly significant correlation

between the two procedures supports the use of the

relatively new continuous flow method for determining AspAT

activity in erythrocytes.

Erythrocyte AspAT activity was reported as

International Units per gram of hemoglobin (Hb) rather than

per liter of red blood cells due to the difficulty in

accurately measuring red cell volume. Hemoglobin

concentration of the hemolysates was measured manually by

the spectrophotometric cyanmethemoglobin method (229). A

commercially available kit was used which provided

methemoglobin prepared from human hemoglobin as the standard

(Sigma Chemical Co. #525-A). Absorbance was determined using

a Spectronic 20 spectrophotometer (Bausch and Lomb Co.).

Plasma PLP

Plasma PLP was determined by the method of Lumeng et

al. (192) as modified by Sloger and Reynolds (191) which is

based on the PLP-dependent decarboxylation of L-tyrosine by

tyrosine apodecarboxylase. The apoenzyme was isolated from

dried Streptococcus faecalis cells grown in

pyridoxine-deficient medium (Sigma Chemical Co. #T4629) and

purified to remove contaminating pyridoxal kinase. All steps

were carried out at temperatures near 00 C. One gram of

cells was suspended in 10 ml cold 0.01 M sodium citrate

buffer, homogenized (Con-Torque homogenizer, Eberbach

Corp.), and centrifuged at 25,000 x g for 15 minutes at 40

C. The supernatant was saved. The pellet was resuspended in

the citrate buffer, and the suspension was sonicated

(Sonifier Cell Disruptor, Model W185, Heat Systems

Ultrasonic, Inc.) for 7 one-minute periods with one-minute

intervals between periods. The centrifugation and

sonification steps were repeated for a total of five times.

The pooled supernatant was slowly brought to 60% saturation

by adding ammonium sulfate (enzyme grade, Sigma Chemical

Co.) with constant stirring. After centrifugation at 25,000

x g at 40 C for 20 minutes, the supernatant slowly was

brought to 85% saturation to precipitate the enzyme fraction

which was resuspended in 5 ml dialysis buffer. The enzyme

suspension was dialyzed (Spectrapor membrane tubing,

12,000-14,000 MW, 25 mm diameter, Fisher Scientific

Products) for 12 hours with one change in dialysis buffer

after 4 hours. The buffer was a solution of 0.3 M sodium

citrate, 24% (v/v) glycerol, and 2mM mercaptoethanol, pH

6.0. The enzyme solution was stored in i-ml vials at -300C.

All steps of the plasma PLP assay were performed in

subdued light to inhibit photolysis of PLP. The plasma

samples were thawed and deproteinized by adding 0.75 ml

saline and 0.25 ml trichloroacetic acid to I ml of plasma.

After mixing with a vortex mixer, the tubes were incubated

at 320 C for 15 minutes and centrifuged at 18,000 x g for 15

minutes at 40 C. After transfer of the supernatant to new

tubes, centrifugation was repeated. The supernatant was

extracted 4 times with water-saturated diethyl ether

(purified, Sigma Chemical Co.) using a 4:1 ether to sample

volume ratio. Residual ether was evaporated with nitrogen

for approximately one hour. The sample extracts were kept on

ice until the assay was begun.

After adding 0.1 ml of 0.1 M sodium citrate buffer

(pH 6.0), 0.2 ml sample extract, and 0.1 ml apoenzyme

(diluted to 0.1 with 0.3 ml sodium citrate buffer) to the

reaction tubes (disposable 16xi00 mm glass tubes, Fisher

Scientific), the reaction mixtures were incubated at room

temperature for 30 minutes for holoenzyme formation to

occur. All samples were analyzed in triplicate. For the

standard curve, 0 0.125 ml of the PLP standard solution

containing 20 ng/ml were substituted for the sample to

provide 0 2.5 ng PLP/tube. The amount of buffer was

adjusted accordingly. The reaction was begun by adding i ml

of labelled tyrosine solution to each reaction tube and

immediately capping the tubes with rubber stoppers fitted

with center wells (Kontes, Inc., #K882320). The wells

contained folded chromatography paper (Whatman #1, Fisher

Scientific) and 0.030 ml of methylbenzonium hydroxide in

methanol (Sigma Chemical Co.) to trap the 1 4 C 0 The

labelled tyrosine solution consisted of 50 Ci

L-14C-tyrosine (56 i Ci/mmol, 97.5% pure, Amersham Corp.),

250 ml 0.1 M sodium citrate buffer (pH 6.0), 31 ml 0.15 N

HC1, and 0.3763 g L-tyrosine. The reaction continued in a

shaking water bath (Thermo-Shake, Forma Scientific Co.) at

320 C and was stopped after exactly 20 minutes by injecting

1 ml of 5 N HCI through the stopper of the tube.

The tubes were kept overnight at room temperature to

ensure complete trapping of the 14CO The center wells were

cut from the stoppers and put into 7-ml polyethylene

scintillation vials. The contents were thoroughly mixed

after adding 3 ml of scintillation fluid (Ready-Soiv NA,

Beckman Instruments). Each sample was counted in a liquid

scintillation counter (6892 Series Liquid Scintillation

System, Tracor Analytic Inc.) for five minutes. The counting

efficiency was 97%. The mean correlation coefficient of all

standard curves of the PLP assay was 0.991 as determined by

linear regression analysis. The coefficient of variation was

7.2 (n=18) for determination of PLP in plasma using this

method. Recovery of PLP added to plasma before

deproteinization was 92%.

Analysis of Vitamin B-6 Supplements

The pyridoxine-HC1 content of the vitamin B-6

supplements used in this study was determined by the reverse

phase high performance liquid chromatrographic (HPLC) method

developed by Gregory and Kirk (42). The vitamin tablets were

manufactured by Hofmann LaRoche Company and consisted of

pyridoxine-HCL, lactose, corn starch, magnesium stearate,

and microcrystalline cellulose. Three tablets of each

concentration were randomly selected from different bottles

for analysis. The tablets were dissolved in 100 ml of

potassium phosphate buffer, pH 2.2. Each solution was

centrifuged at 1000 x g, and the supernatant was filtered

through a 0.45 I'm filter before analysis. The HPLC system

consisted of an Altex model 312 chromatograph and

octadecylsilica column (Partisil 10 ODS-3, Whatman, Inc.).

The column was equilibriated with the mobile phase, 0.033 M

potassium phosphate pH 2.2, for 30 minutes before use.

Pyridoxine was measured by an Altex ultraviolet analytical

detector at 280 nm, and results were recorded on a strip

chart recorder. Calibration standards of 25 250 i g /ml

pyridoxine-HC1 (Sigma Chemical Co.) were used. The measured

mean pyridoxine-HCl content of the tablets containing 0,

2.6, 5, 7.5, 10, 12.5, 15, and 20 mg was 0, 2.5, 4.9, 7.3,

9.2, 12.1, 14.8, and 20.0 mg.

Dietary Analysis

The dietary information obtained from the 24-hour

dietary recalls was analyzed by computer using the Nutrient

Dietary Analysis System at Southern Illinois University,

Carbondale, Illinois (230). This system was used by the

Research Triangle Institute to estimate the dietary intake

of pregnant women in a national evaluation of the Special

Supplemental Food Program for Women, Infants and Children


Statistical Analysis

Vitamin B-6 supplementation levels were plotted against

plasma PLP levels and erythrocyte AspAT activity before and

after stimulation by exogenous PLP at 30 weeks gestation and

at delivery to provide dose response curves. Analysis of

variance (ANOVA) procedures were used to examine the effects

of vitamin B-6 supplementation on the various biochemical

measurements of vitamin B-6 status of mothers at 30 weeks

gestation and at delivery and of the fetus at delivery

(231). The effects of such factors as race, parity, tobacco

and alcohol use on vitamin B-6 status were also tested by

ANOVA. The effect of vitamin B-6 supplementation on infant

condition at birth as measured by birth weight, birth length

and placental weight were determined by ANOVA procedures.

ANOVA was also used to test the effect of vitamin B-6

supplementation level on percent changes in biochemical

measurements of maternal vitamin B-6 status between the

initial visit and 30 weeks gestation and between the initial

appointment and delivery. The Kruskall-Wallis one-way

analysis of variance by ranks test was used to examine the

effect of vitamin B-6 supplementation on the Apgar scores of

infants at 1 and 5 minutes after birth.

Analysis of covariance was used to test the effect of

vitamin B-6 supplementation and total vitamin B-6 intake on

measurements of vitamin B-6 status at 30 weeks gestation and

delivery while accounting for vitamin B-6 status at the

first clinic visit (231). Differences in the various

indicators of vitamin B-6 status and measurements of

pregnancy outcome between subjects who experienced morning

sickness and those who did not were tested by the Student's

t test (231). This test was also used to compare indicators

of maternal vitamin B-6 status measurements of pregnancy

outcome between mothers consuming 7.5 mg/day and more

supplemental vitamin B-6 and those taking 5 mg and less.

Differences in these parameters between maternal and cord

plasma PLP levels above and below the 50th percentile were

tested by the Student's t test.

Linear regression methods were used to determine the

relationships between plasma PLP levels, erythrocyte AspAT

activity and stimulation by exogenous PLP at each sampling

time (231). The correlation between maternal and cord plasma

PLP levels was also examined by linear regression. Linear

regression methods were used to assess the relationship

between measurements of vitamin B-6 status and measurements

of infant condition at birth. Possible relationships between

such factors as age, weight, prepregnancy weight, degree of

morning sickness, tobacco and alcohol use and measures of

infant condition at birth were also tested by regression

analysis. Linear regression methods were used to examine the

relationship between the degree of morning sickness

experienced by subjects during the first trimester and

measurements of vitamin B-6 status, measurements of

condition of the infants at birth, parity, age, weight,

tobacco and alcohol use. All statistical analyses were

performed by computer using the Statistical Analysis System



Nonpregnant Group

The mean values of the biochemical indicators of

vitamin B-6 status in twenty-six healthy nonpregnant young

women provided "control values" with which to compare the

pregnant experimental group of women (table 3). These women

did not take vitamin B-6 supplements and had no history of

oral contraceptive use. The fasting plasma PLP level was

59.0 + 19.0 pmol/ml (mean+SD), or 14.6 + 4.7 ng/ml. Plasma

PLP values ranged from 28.7 to 117.4 pmol/ml. Mean plasma

PLP levels reported in the literature include 37.6 pmol/ml

(233), 42.4 pmol/ml (190), and 68.3 pmol/ml (224) for

nonpregnant women and 51.8 pmol/ml (233) and 59.8 pmol/ml

(234) for men. AspAT activity was 2.08 + 0.62 U/g hemoglobin

and ranged from 1.30 to 3.79 U/g. The mean percent

stimulation of AspAT activity by exogenous PLP was 21.6 +

15.7% with a 0 47.8% range.

Linear regression analysis revealed no correlation

between plasma PLP values and erythrocyte AspAT activity or

stimulation by added PLP. A negative correlation (p<.0005)

was found between AspAT activity and percent stimulation by

Table 3. Biochemical measurements of vitamin B-6 status of
nonpregnant comparison group and pregnant subjects

Nonpregnant Women Pregnant Subjects

(n=26) (n= 19 6)

Plasma PLP 59.0+19.0 37.1+ 25.3*

Erythrocyte AspAT 2.08+0.62 2.6 3+1 .57
activity (U/g Hb)
** _**
Stimulation of 22+16 86+140
AspAT by PLP (%)

Means with asterisks are significantly different (p<0.0001).

exogenous PLP which indicates that lower AspAT activity was

associated with insufficient coenzyme.

Since plasma PLP is considered to be the most sensitive

indicator of vitamin B-6 status, this measurement was used

to compare the pregnant women with the nonpregnant group.

The distribution curve of these plasma PLP values was highly

skewed to the right skewnesss = 1.15); therefore, plasma PLP

values were converted to their base 10 logarithm to

normalize the distribution curve (235). The mean and

standard deviation of the log values were determined. The

lowest normal value for the nonpregnant group was defined as

the mean minus 2 standard deviations which was 31.5 pmol/ml

(7.8 ng/ml). Lumeng et al. (190) used the same procedure to

normalize data by log conversion of plasma PLP values prior

to calculating the mean minus 2 standard deviations. The

value reported by Lumeng et al. (190) using this technique

with a group of nonpregnant women was 19.0 pmol/ml (4.7

ng/ml) compared with 31.5 pnol/ml (7.8 ng/ml) in the present


Pregnant Group

Description of Subjects

The 196 pregnant women who served as subjects in this

study ranged in age from 17 to 38 years. Fifty-two percent

of the group was white and 48% black. The mean gestational

age at the initial prenatal clinic visit was 15 + 4 weeks

and ranged from 6 to 21 weeks. Twenty-two percent of the

women had no previous pregnancies, 27% had 1, 26% had 2, and

25% had 3 to 7 previous pregnancies. Fifty-three percent

reported no tobacco use at the time of the initial

appointment, and 85% claimed no consumption of alcoholic

beverages. The mean hematocrit for the group was 37 + 3%,

and values ranged from 30 to 49%.

Nutrient Intake

Nutrient intakes were computed for 65 subjects from

24-hour dietary recalls at or near 30 weeks gestation (table

4). The mean dietary vitamin B-6 intake was 1.43 + 1.28

mg/day (mean + SD) which is 55% of the 1980 RDA for vitamin

B-6. Eighty-three percent of the subjects consumed less than

the RDA.

The mean daily energy intake was 2152 + 843 kcal which

represents 94% of the RDA for women over 23 years old.

Values ranged from 885 to 4764 kcal. The mean protein intake

was 81.7 + 38.6 g (122% of the RDA for women over 19 years)

which ranged from 21 to 197 g. When expressed as ratios of

the vitamin to protein and energy intake, the mean vitamin

B-6 intake was 18.9 + 19.3 Pg/g protein and 0.67 + 0.63 mg/

1000 kcal.

Vitamin B-6 Status at Initial Clinic Visit

The means of the biochemical measurements of vitamin

B-6 status for the 196 pregnant subjects determined at the

initial clinic visit are compared with the mean values of

the nonpregnant group in table 3. The mean plasma PLP level

Table 4. Nutrient intakes of pregnant subjects
24-hour dietary recalls.

computed from

Nutrient Intake Mean intake
(mean+SD) as % RDA

Vitamin B-6



Vitamin B-6/protein

Vitamin B-6/energy

Vitamin A

Vitamin D

Vitamin E

Vitamin C





Vitamin B-12






1.43+1.28 mg

81.7+38.6 g

2152+843 kcal

18.9+19.3 9g/g

0.67+0.63 g/kcal

1333.9+1545.2 Ig

5.79+5.63 9 g

13.6+12.5 mg

162.0+142.9 mg

329.0+267.0 L g

22.8+16.3 mg

2.53+1.03 mg

1.61+1.07 mg

3.9+3.6 I'g

963.5+683.2 mg

1181.0+653.3 mg

17.9+13.4 mg

224.0+149.4 mg

9.2+5.4 mg



















of the pregnant subjects at the initial prenatal appointment

was 37.1 + 25.3 pmol/ml (9.2 + 6.3 ng/ml, n=196) which,

although significantly lower (p<0.0001) than the mean of the

nonpregnant comparison group, is not considered abnormally

low as defined previously (<31.5 pmol/ml). However, 44% of

the group did have plasma PLP levels below 31.5 pmol/ml. The

mean erythrocyte AspAT activity for the group was 2.63 +

1.57 U/g hemoglobin (n=178), and mean percent stimulation

with exogenous PLP was 86% + 140%.

Parity, race, tobacco smoking, and alcohol use had no

effect (p>0.05) on plasma PLP levels, erythrocyte AspAT

activity or stimulation by added PLP measured at the initial

prenatal visit. When subjects were categorized as "low" or

"adequate" on the basis of plasma PLP levels greater or less

than 31.5 pmol/ml at the initial clinic visit, there were no

differences between groups in age, weight, prepregnancy

weight, parity, blood pressure, degree of morning sickness,

and hematocrits determined at the initial visit. No

significant differences between these groups were found in

maternal plasma PLP levels at 30 weeks gestation or

delivery, cord plasma PLP levels, birth weight, birth

length, placental weight, or Apgar scores at 1 or 5 minutes

after birth. The mean gestational age (stage of pregnancy)

of subjects in the low plasma PLP group was significantly

higher (16 + 4 weeks) than those with adequate plasma PLP

levels at the initial appointment (14 + 4 weeks, p<0.02).

Morning sickness occurs in the majority of pregnant

women (210), and the relative vitamin B-6 deficiency

observed during pregnancy has led to an interest in the

relationship between vitamin B-6 status and morning

sickness. Therefore, efforts were made to detect any

relationship between vitamin B-6 status and the degree of

morning sickness experienced in early pregnancy by the

subjects in this study. At the initial prenatal visit

subjects were questioned regarding their morning sickness

experiences, and the following categories were used to rank

the degree of morning sickness: i) none, 2) mild nausea, 3)

occasional nausea and vomiting, and 4) daily nausea and

vomiting. These findings are summarized in table 5.

Thirty-three percent of the subjects were in category 1, 30%

in 2, 22% in 3, and 15% in 4. There was no correlation

between any measurements of vitamin B-6 status at the

initial clinic visit and the degree of morning sickness

experienced by this group in early pregnancy. Women who

experienced morning sickness had a greater number of

previous pregnancies (2.0 + 1.6) than those who had no

morning sickness symptoms (1.4 + 1.4, p<0.05). The placental

weights of mothers who had morning sickness were higher (638

+ 123 g) than those who did not (517 + 94 g, p<0.002). There

were no differences in the following parameters between

women who did and did not experience morning sickness:

plasma PLP levels, erythrocyte AspAT activity or stimulation

by PLP, age, weight, prepregnancy weight, blood pressure,

Table 5. Comparison of various measurements of vitamin B-6
status and infant condition at birth and parity between
pregnant women who did (Group 1) or did not (Group 2)
experience morning sickness in early pregnancy.

Group i

Plasma PLP

AspAT activity
(U/g Hb)

Stimulation of
AspAT activity

Parity (No.)

Birth weight

Birth length

Apgar score,
i min

Apgar score,
5 min

Placenta weight

(n= 119)



2.0+1.7 *






Group 2


2.3 9 + 4


1.4+1.4 *



8.3+ 1 .0


517+94 **
(n= 18)


p< 0.002

Means with asterisks are si g n if i cant 1 y different.



Means with asterisks are

hematocrit, tobacco or alcohol use, birth weight, birth

length, or Apgar scores at i and 5 minutes after birth.

Effect of Vitamin B-6 Supplementation on Vitamin B-6 Status

As described in the Experimental Methods section, blood

samples were obtained from 71 subjects participating in the

study at 30 weeks gestation, and maternal and cord blood

samples were obtained from 40 subjects at delivery. Only

those subjects who complied with the supplementation

protocol as determined by questioning the subjects, counting

remaining vitamin B-6 tablets at the 30 week appointment,

and obtaining information from the attending nurses and

doctors were included in final analysis of the data. This

resulted in a final total of 46 subjects at 30 weeks

gestation and 22 at delivery. Sixteen of these subjects

provided both 30-week and delivery samples. The final number

of subjects included within each supplementation group

varied according to the parameter measured and is indicated

in tables by the designation "n". The 12.5 mg vitamin B-6

supplementation group was excluded from final data analysis

because an insufficient number of subjects complied with

supplementation protocol (n=3 at 30 week appointment and n=l

at delivery). The total daily vitamin B-6 intake for the

subjects was determined by adding the total dietary vitamin

B-6 intake of the entire experimental group (1.43 mg/day) to

each vitamin B-6 supplement level. The limitations of the

24-hour dietary recall method do not permit using the

estimate of the dietary intake of each supplementation group

since the sample numbers are too small (156, 157).

The mean values of the biochemical indicators of

vitamin B-6 status for each supplementation group are shown

in table 6. The biochemical measurements at the initial

prenatal clinic visit were not significantly different

(p>0.05) among supplementation groups. In this study,

erythrocyte AspAT activity and stimulation by exogenous PLP

did not correlate with vitamin B-6 supplementation level and

were not linearly responsive to graded levels of vitamin B-6

intake. Analysis of covariance with the initial visit AspAT

activity and stimulation by PLP as covariables indicated

that there was no significant difference in these two

measurements among supplementation groups at 30 weeks


Maternal plasma PLP levels, however, were positively

correlated with vitamin B-6 supplementation at 30 weeks

gestation (r=0.55, p<0.0005) and at delivery (r=0.54,

p<0.01). Analysis of covariance with initial visit plasma

PLP levels as the covariable indicate that vitamin B-6

supplementation significantly affected plasma PLP levels.

There were no significant differences in maternal plasma PLP

levels at delivery or in cord plasma PLP levels among

supplementation groups. The linear responses of maternal

plasma PLP levels at 30 weeks gestation (r=0.54, p<0.0001)

and delivery (r= 0 .55, p< 0.00 7) to vitamin B -6

supplementation levels are shown in figures 5 and 6,

0 .
C'4 r























c-o -







E '
0 0

in > i
E a -

to C .C.C (n C
-- S inS! S
CL4 43


n C


+I c

+. 4





0 C


0 II

+I c







+1 II


IN c


+1 II

0 -



+1 +1M

+1 +l I


+1 +!

IN 0
+1 +I




0- i

*- 0

+1 +I|r

0, 0u,

r- c-i
+1 +1
1.5 LaS II

U >

E =-
.' 'CL

I n0


+1 i








- C









n E
u 8 a
C -.

E3~.E as

in0 -I.- vE

td '-0u


4 -





. +bi

CO +1
-4 C IlI

r'I co .





0 0 0


c n
.el *

0+ *


01 ,4 0
S1 II 'I 0
P: t"t 0

0 00

00 0

0 0




S on

a >




E a E-
a. aI


a t E

0 >

E 4*9

(A c

(TW1lowd) dcld vwsTeld


- -4

0 +
.4 ...-

o 0 0
cn oC'

pa La.

o a













4i +-
S0 c




+ c


0 0


0 0 0

(luI/Iouwd) dild eusel.d

respectively. The response of cord plasma PLP levels to

vitamin B-6 supplementation, however, was not linear, as can

be seen in figure 7. The cord plasma PLP level increases

rapidly until about the 7.5 mg supplementation level when

the response leveled off.

The effect vitamin B-6 supplementation on the maternal

plasma PLP levels over time and on cord plasma PLP is shown

in figure 8. In order to compare the effects of the various

levels of supplementation more easily, figure 9 represents

the same bar graph after all PLP levels have been adjusted

in order that all initial maternal plasma PLP levels are

equal. It is apparent that response of maternal plasma PLP

at 30 weeks gestation and delivery and cord plasma PLP is

essentially the same whether the mother is taking a placebo

(0 mg) or 2.6 mg of supplemental pyridoxine-HCI. Neither of

these levels is able to prevent a decrease of approximately

30% in maternal plasma PLP levels at 30 weeks gestation, and

mean maternal plasma PLP levels at delivery are less than

half the levels present at the initial clinic visit in early

pregnancy in both supplementation groups. In contrast, all

other supplementation levels (5 mg and above) were able to

increase the 30 week plasma PLP levels above the initial

values. Supplementation with 5 and 7.5 mg of pyridoxine-HC1

resulted in an increase of approximately 25% while 10 and 15

mg resulted in an increase of approximately 100%. The most

dramatic effect was by the 20 mg supplement which resulted

0 0

0 0

0 0 0 0
0 s 0 m
Cd 1 O

(Tui/ouwd) did eLusWid


0 -




01 -





0 .-
a -

0 ..
A >


+. E




o *
0 L

> w=
o a d

> 5.


o L I
E m (10
* BcuD








01 >















0 0.


(luw/Ioud) dId ,1,wsvld






in LA

d E

o E E















"u l

E nj
> n 0W


oCn 0



o 0o o 0
0 m O
N C(lui/Ioud) eInd wuus id



in a four-fold increase of 30 week plasma PLP levels above

initial values.

Five milligrams of supplemental pyridoxine-HCI or 5.5

mg total maternal intake of vitamin B-6 (diet plus

supplement as pyridoxine equivalents) was able to maintain

the 30-week plasma PLP at a level comparable to initial

values. However, 5 mg of supplemental pyridoxine-HCI did not

prevent a decrease of more than 60% in maternal plasma PL?

level at delivery from the initial values. The 7.5 mg and

higher supplemental levels were able to prevent this

decrease. This finding corresponds with the observation that

fetal plasma PLP levels appeared to saturate at 7.5 mg

supplemental pyridoxine-HC1 (f-igure 7).

Changes in the biochemical measurements of vitamin B-6

status within each subject between the initial visit and 30

weeks gestation and/or delivery were calculated as

percentages, and the means within each supplementation group

are shown in table 7. Analysis of variance procedures

indicated that vitamin B-6 supplementation had a significant

effect (p<0.01) on the change in percent stimulation of

erythrocyte activity by exogenous PLP between the first

visit and 30 weeks gestation (figure 10). However, no

statistically significant effect was found in changes in

erythrocyte AspAT activity between the first visit and 30

weeks gestation. Vitamin B-6 supplementation significantly

affected the percent changes in maternal plasma PLP levels

+ It



No 1
- "

.1 -

- -


a o



a -0




au a
u 0

f- XI

E a






5: >"



&< U



4, II






m c


* 11

M ,










o i











c I

l c


o c


















o -0


a r- 4 4 C


I +



1-4 .5;
< 4 -
in E
42 <




Sn E







o c


+ 0
u CL
,d ,.

- a



" c.


u 0

. '

. ..

. 3
asc 0
54 LA


0 0 1 0
+ + + I
a6uvL4p .uajajad


between the first and 30 weeks gestation (p<0.01) but not

between the first visit and delivery (figure 11).

When plasma PLP levels measured at 30 weeks gestation

were classified as "low" (<31.5 pmol/ml) or "adequate"

(31.5 pmol/ml), there were no significant differences

between these groups in hematocrit, maternal or cord plasma

PLP levels at delivery, or measures of pregnancy outcome. At

delivery, fetal plasma PLP levels were significantly lower

from mothers who had low levels of plasma PLP than fetal

levels from mothers with adequate PLP levels (83.5 versus

154.4 pmol/ml, respectively; p<0.01). Maternal and fetal

plasma PLP levels at delivery were positively correlated

(r=0.52, p<0.002). This relationship has been reported

previously and confirms that fetal dependence upon the

mother for vitamin B-6. Analysis of covariance with initial

maternal plasma PLP levels as the covariable showed no

differences in ratios of maternal and cord plasma levels

among the supplementation groups.

Analysis of variance procedures revealed no differences

in measurements of infant condition at birth which included

birth weight and length, placental weight, and Apgar scores

at i and 5 minutes after birth among the supplementation

groups. Since 7.5 mg of maternal pyridoxine-HCl

supplementation during pregnancy appeared to saturate the

plasma PLP of the fetal-placental unit, the condition of

infants at birth whose mothers took 7.5 mg or more was

compared with that of infants whose mothers took 5 mg or

o G
... >

C '

0 m
3 n

D '

0 0 0 o 0
0 O O O
+ + +
a6ueq3 4.ua3Jad




> >0
w a-

In n

01 01 L

E d

E no




* 4
*r 01

4.- 4L *
IrI4 .

less supplemental pyridoxine-HC1. The means of the

indicators of birth outcome and vitamin B-6 status at

delivery are shown in table 8. Both maternal and cord

plasma PLP levels were significantly higher (p<0.005) when

supplementation was 7.5 mg. Apgar scores at 1 minute after

birth were significantly higher (p<0.05) for infants whose

mothers took 7.5 mg or more supplemental pyridoxine-HC1 than

for infants of mothers who took 5 mg or less. No effect of

maternal vitamin B-6 status at 30 weeks gestation or

delivery on infant condition at birth was demonstrated when

subjects were categorized as having "low" or "adequate"

plasma PLP levels. The means of the measurements of the

infant condition at birth for maternal plasma PLP levels at

delivery and cord plasma PLP levels grouped above or below

the 50th percentile are presented in table 9. The infant

Apgar scores at 1 minute after birth were significantly

higher (p<0.01) when maternal plasma PLP levels at delivery

and cord plasma PLP levels were in the upper 50th

percentile. There were no significant differences between

the two groups in birth weight, birth length, and placental

weights. Other factors such as race, parity, alcohol or

tobacco use did not affect the condition of the infant at

birth. The frequency of specific Apgar scores in each

supplementation group is shown in table 10.

Incidence of Gestational Diabetes and Pre-eclampsia

Only three of the original enrollees in the study

developed gestational diabetes during the course of

Table 8. Measurements of vitamin B-6 status at delivery anc
infant condition at birth when daily vitamin B-6
supplementation was (5mg or >7.5 mg.

Vitamin B-6
0-5 mg/day

Supplementation Level
7.5-20 mg/day

Maternal plasma PLP
at delivery (pmol/ml)

Cord plasma PLP



(n= 12)

p < 0 0 5

152.7+54.3 p < 0.005

Birth weight

Birth length


(n= 16)

Placenta weight

Apgar score,
i min

Apgar score,
5 min

(n= i)




(n= 17)

548 +1 14

8.6+0.6 *

9. 1_+0.6

Means with asterisks are significantly different.

p < 0 5

Means with asterisks

are significantly different.












E .,
* 01



E "

01 -I

L 0

c >





in '- +1
*a! S0;

EU -. ^

P- 0
^ E

l o

c. .cD




< +1
01 C


00 It

+r c
- c











a 0=





+ 10


1 .
0 i


U* C





+1 a




+| II

0 C

0 -'

0I "







+ c

Table 10. Number of infants receiving a given Apgar score at
1 minute after birth in each maternal vitamin B-6
supplementation group.

Vitamin B-6 Supplementation

Level (mg/day)

0 2.6 5 7.5 10 15 2 0

Apgar score

9 5

1 1

2 0

1 0

2 3

4 2

0 0 0 0

0 0 0 0

1 0 0 1

1 1 1 1

5 3 5 3

pregnancy. No delivery samples were available for these

subjects, although 30 week blood samples were obtained.

However, these subjects were not included in statistical

analyses involving the effect of vitamin B-6 supplementation

since their compliance with the protocol was questionable.

The 30-week plasma PLP levels for these women who were given

5, 10, and 15 mg pyridoxine-HC1 supplements were 18.2, 73.1,

and 60.6 pmol/ml, respectively. All three delivered

normal-weight infants (>2500 g). The Apgar scores of the

infant of only one mother were low. This subject had the

lowest plasma PLP level at 30 weeks gestation of this group

(18.2 pmol/ml) although her plasma PLP level at the initial

visit was the highest (49.7 pmol/ml). The Apgar scores at I

and 5 minutes after birth for her infant were i and 6.

Ten subjects of the original study volunteers were

diagnosed as having pre-eclampsia at term. Blood samples

were obtained at delivery from four of these subjects. Two

subjects took the 20 mg pyridoxine-HC1 supplement during

pregnancy, and their maternal plasma PLP levels were 29.5

and 22.2 pmol/ml. The cord plasma PLP levels of these

subjects were both 99.8 pmol/ml. These values are among the

lowest in the 20 mg supplementation group. The other two

subjects were given 12.5 mg supplements; but their

compliance with the protocol was questionable, and they were

therefore not included in the statistical analysis of the

data. Both had low plasma PLP levels at delivery, and one

was the lowest of any subject in the study. The maternal

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E3NGFH0BC_M8ZW9O INGEST_TIME 2012-02-17T16:30:28Z PACKAGE AA00003439_00001