Nutrient Requirements

Assessment of Vitamin B-6 Status in Young Women Consuming a

Controlled Diet Containing Four Levels of Vitamin B-6 Provides an
Estimated Average Requirement and Recommended Dietary Allowance1,2
Christine M. Hansen, Terry D. Shultz,3 Ho-Kyung Kwak,* H. Sara Memon
and James E. Leklem*
Department of Food Science and Human Nutrition, Washington State University, Pullman, WA 99164-6376
and *Department of Nutrition and Food Management, Oregon State University, Corvallis, OR 97331-5103

ABSTRACT The Recommended Dietary Allowance (RDA) of vitamin B-6 for young women was recently reduced
from 1.6 to 1.3 mg/d based on an adequate plasma pyridoxal phosphate (PLP) concentration of 20 nmol/L. To
assess vitamin B-6 requirements and suggest recommendations for intake, seven healthy young women con-
sumed a controlled diet providing 1.2 g protein/kg body weight for a 7-d adjustment period (1.0 mg vitamin B-6/d)
and three successive 14-d experimental periods (1.5, 2.1 and 2.7 mg/d, respectively). Direct and indirect vitamin

Downloaded from jn.nutrition.org by guest on August 9, 2017

B-6 status indicators were measured in plasma, erythrocytes and urine. Indicators most strongly correlated with
vitamin B-6 intake [i.e., plasma and erythrocyte PLP, urinary 4-pyridoxic acid (4-PA) and total vitamin B-6] were
regressed on vitamin B-6 intake and the dietary vitamin B-6 to protein ratio. Inverse prediction using adequate and
baseline values estimated vitamin B-6 requirement. Adequate values were determined for plasma PLP and urinary
4-PA from baseline values of 60 previous subjects, using the statistical method suggested by Sauberlich. The
current study suggests a vitamin B-6 Estimated Average Requirement (EAR) for young women of 1.1 mg/d or 0.016
mg/g protein, and a RDA of 1.5 mg/d or 0.020 mg/g protein. When results from this study are combined with data
from four other recent studies, the combined data predict an EAR of 1.2 mg/d or 0.015 mg/g protein, and a RDA
of 1.7 mg/d or 0.018 mg/g protein. This study suggests that the current vitamin B-6 RDA may not be adequate. J.
Nutr. 131: 1777–1786, 2001.

KEY WORDS: ● vitamin B-6 ● status assessment ● Estimated Average Requirement

● Recommended Dietary Allowance ● women

In the 1998 report of the Standing Committee on the
Scientific Evaluation of Dietary Reference Intakes (DRI),4 the
Recommended Dietary Allowance (RDA) of vitamin B-6 for
adult women was reduced from 1.6 to 1.3 mg/d (1). The
Committee based their recommendation primarily on data
from six studies (2–7). Data from these studies were reevalu-
ated to determine the vitamin B-6 intake required for a plasma
pyridoxal phosphate (PLP) concentration of 20 nmol/L.
Plasma PLP concentration was chosen as the standard for
1
Presented in part at Experimental Biology 99 (Washington, DC) and 2000
(San Diego, CA) [Kwak, H. K., Hansen, C., Hardin, K., Ridlington, J., Leklem, J. E.
& Shultz, T. D. (1999) A positive effect of vitamin B-6 on the immune response
of young women. FASEB J. 13: A699 (abs.); Hansen, C., Kwak, H. K., Memon,
H. S., Shultz, T. & Leklem, J. (1999) Plasma total homocysteine concentrations
in women consuming four levels of vitamin B-6 intake. FASEB J. 13:A889 (abs.);
Shultz, T. D., Hansen, C. M., Memon, H. S., Kwak, H. K. & Leklem, J. E. (2000)
Urinary and erythrocyte vitamin B-6 (B6) status indicators of women consuming
four levels of B6 intake. FASEB J. 14: A242 (abs.)].
2
Supported by U.S. Department of Agriculture NRICGP grant #97–35200 –
4238.
3
To whom correspondence should be addressed. E-mail: shultz@wsu.edu
4
Abbreviations used: DRI, Dietary Reference Intake; EALT, erythrocyte ala-
nine aminotransferase; EAR, Estimated Average Requirement; EAST, erythrocyte
aspartate aminotransferase; 4-PA, 4-pyridoxic acid ; PL, pyridoxal; PLP, pyridoxal
phosphate; PMP, pyridoxamine phosphate; PN, pyridoxine; PNG, PN glucoside;
RDA, Recommended Dietary Allowance.
adequate status because this measure appears to reflect tissue
stores (8). In the absence of evidence linking a particular
concentration to favorable or unfavorable health outcomes, a
concentration of 20 nmol/L was chosen as the standard for
adequate status. The DRI Committee stated that a plasma PLP
concentration of 20 nmol/L is “not accompanied by observable
health risks” (1) and was suggested by one group of investiga-
tors (8,9).
The study reported here was undertaken to investigate the
relationship between vitamin B-6 intake and measures of
immune function to determine whether these measures could
be used to establish vitamin B-6 requirements. The immune
function results will be published separately. In this paper,
recommendations for vitamin B-6 intake will be assessed on
the basis of the effect of vitamin B-6 intake and the dietary
vitamin B-6 to protein ratio on several vitamin B-6 status
indicators in plasma, erythrocytes and urine of young women.
Following DRI Committee methodology, the Estimated Aver-
age Requirement (EAR) will be determined and the RDA
calculated. In addition, data from this study and several other
recent studies (3,5–7) will be combined and recommendations
for vitamin B-6 intake proposed on the basis of the combined
data.

0022-3166/01 $3.00 © 2001 American Society for Nutritional Sciences.

Manuscript received 17 November 2000. Initial review completed 4 January 2001. Revision accepted 8 March 2001.

1777
1778 HANSEN ET AL.

TABLE 1
Descriptive characteristics of women who consumed four levels of vitamin B-6 for a 7-d adjustment period and
three 14-d experimental periods

Weight

Subjects Age Height Initial Final BMI1 Race

y cm kg kg/m2

1 37 161 64.0 62.0 24.7 Caucasian

2 21 156 67.0 68.6 27.5 Caucasian
3 29 167 65.3 64.4 23.4 Caucasian
4 28 159 52.1 50.9 20.6 Asian
5 31 164 55.5 54.1 20.6 Asian
6 32 158 55.5 53.0 22.2 Asian
7 21 162 54.5 53.8 20.8 Caucasian
Mean ⫾ SD 28 ⫾ 6 161 ⫾ 4 59.1 ⫾ 6.0 58.1 ⫾ 6.3 22.8 ⫾ 2.6

1 Initial body mass index.

SUBJECTS AND METHODS diets previously described by Hansen et al. (5) and Huang et al. (6).

Downloaded from jn.nutrition.org by guest on August 9, 2017

Subjects. Premenopausal women (n ⫽ 8) were recruited from the
Washington State University community. Potential subjects com-
pleted a health history questionnaire, gave a fasting blood sample for
clinical chemistry evaluation, underwent xylose absorption testing
(10) and were examined by the study physician. Subjects were in
good health, nonsmokers and not taking prescription medications or
nutritional supplements. Three-day dietary records were obtained
from each subject 4 wk before the beginning of the study to evaluate
subjects’ usual nutrient intakes. Subjects’ characteristics are listed in
Table 1. The screening and experimental procedures were reviewed
and approved by the Institutional Review Board of Washington State
University, and informed consent was obtained from each subject.
Subjects were housed in the Human Metabolic Unit at Washington
State University throughout the 49-d study and were instructed to
maintain their usual activity level.
Experimental design. The protocol consisted of a 7-d adjustment
period and three successive 14-d experimental periods (Table 2).
Subjects consumed a nonvegetarian, 3-d rotating menu providing 1
mg/d (5.91 ␮mol/d) vitamin B-6 and 1.2 g protein/kg body weight.
After the 7-d adjustment period, vitamin B-6 was supplemented as a
pyridoxine (PN) hydrochloride (Sigma Chemical, St. Louis, MO)
solution given at breakfast. Total vitamin B-6 intake for the three
14-d experimental periods (diet plus supplement) was 1.5, 2.1 and 2.7
mg/d (8.86, 12.41 and 15.95 ␮mol/d), respectively.
Diet. The basal diet fed throughout the four periods (Table 3)
consisted of natural plant and animal foods, and was adapted from
TABLE 2
Experimental design of study during which seven women
consumed four levels of vitamin B-6 for a 7-d adjustment
period and three 14-d experimental periods
Ratio of vitamin B-6
Metabolic periods Vitamin B-6 intake to protein
mg/d (␮mol/d) mg/g
Screening 1.4 ⫾ 0.6 (8.3 ⫾ 2.4)1,2 0.020 ⫾ 0.007
Adjustment (7 d) 1.0 (5.91)3 0.014 ⫾ 0.001
Period 1 (14 d) 1.5 (8.86)3 0.021 ⫾ 0.002
Period 2 (14 d) 2.1 (12.41)3 0.029 ⫾ 0.003
Period 3 (14 d) 2.7 (15.95)3 0.037 ⫾ 0.004
1 Mean ⫾ SD.
2 Vitamin B-6 intake calculated from 3-d diet records.
3 Vitamin B-6 intake determined by microbiological assay.
Average daily nutrient composition of the 3-d rotating menu is
detailed in Table 4. Meals were prepared in the kitchen of the
Metabolic Unit and eaten in the Unit’s dining room, except for
occasional take-out lunches. Food portions and recipe ingredients
were weighed accurately to within 0.1 g, and recipes and cooking
times were standardized. Food items, except perishables, were pur-
chased in a single lot to minimize variability in nutrient composition.
Subjects consumed only those foods and beverages prepared for them
or permitted. Egg white powder (reconstituted and cooked) was added
to menu items to adjust protein intake according to individual body
weight. To provide the 1989 RDA (11) for all nutrients except
vitamin B-6, subjects were supplemented with 4 mg nicotinic acid
and 200 ␮g folic acid (Sigma Chemical), 8.8 mg elemental iron
(Feosol; Smith Kline Consumer Products, Philadelphia, PA), 333 mg
calcium, 133 mg magnesium and 5 mg zinc (Cal-Mag Zinc; Thrifty
PayLess, Wilsonville, OR). Additional energy sources that contained
very little or no vitamin B-6 [i.e., sugar, margarine, jam, soft drinks,
hard candy and cookies (providing ⬍0.01 mg/d)] were offered as
required to maintain body weight, and intakes of these foods were
recorded daily. Tea and instant coffee were provided at an amount
selected by each subject at the beginning of the study as their usual
daily consumption.
Sample collection and analysis. Food composites of the basal
diet were made on each of the three different menu days several times
during the study. Total vitamin B-6 was determined in the composites
by microbiological assay (12) and PN glucoside (PNG) by the
method of Kabir et al. (13).
Daily 24-h urine collections were obtained, using toluene as a
preservative, throughout the study. Aliquots of urine were stored at
⫺20°C until analysis. Urinary creatinine was assayed to determine
completeness of collection (14). Urinary 4-pyridoxic acid (4-PA) was
analyzed by an HPLC procedure (15) omitting the acidification step.
Mean (⫾SD) recovery of added 4-PA was 98 ⫾ 14%. Total vitamin
B-6 in urine was determined by a microbiological assay (16). Urine
was tested weekly for glucose, bilirubin, ketones, blood, protein, pH
(Bili-Labstix; Bayer Corporation Diagnostics Division, Elkhart, IN)
and pregnancy (QuPID; Stanbio Laboratory, San Antonio, TX).
Weekly blood samples were collected into EDTA and heparin
anticoagulated Vacutainer (Becton Dickinson, Rutherford, NJ) tubes
after an overnight fast, and immediately placed on ice. After whole
blood was removed for hematology determinations (T660 Coulter
Counter; Coulter Electronics, Hialeah, FL), plasma and erythrocytes
were separated by centrifugation (1430 ⫻ g at 4°C). Plasma was
removed and aliquots stored at ⫺40°C. Erythrocytes were washed
three times with saline, an aliquot of packed cells was removed for
assay of alanine and aspartate aminotransferase activities and the
remainder was frozen at ⫺40°C. Lymphocytes were separated from
one tube of heparinized blood for immune function tests and deter-
 VITAMIN B-6 REQUIREMENT AND RECOMMENDATIONS 1779

TABLE 3
Basal diet consumed by women during the adjustment and experimental periods

Breakfast Lunch Dinner Other foods1

g g g g

Day 1 Grape juice2 180 Apple, cored 70 Milk, 2% milkfat 250 Margarine 30
Milk, 2% milkfat 180 Bread, French 25 Dinner roll 28 Popcorn 15
Scrambled egg Salad Fruit cocktail 100
substitute 75 Lettuce 50 Vegetable goulash
Bread, wheat 25 Carrot 20 Egg white powder3
Red cabbage 20 Tofu 60
Kidney beans,
Celery 30 canned 30
Kidney beans, canned 30 Frozen carrots 35
Cheddar cheese 35 Celery 35
French dressing 20 Condensed tomato
soup 42
Egg noodles, cooked 100
Onion flakes 1.5
Spike seasoning4 1.0
Day 2 Orange juice2 180 Apple juice2 180 Milk, 2% milkfat 250 Margarine 30
Milk, 2% milkfat 180 Sandwich Peaches, canned 100 Graham crackers 40

Downloaded from jn.nutrition.org by guest on August 9, 2017

Puffed wheat 18 Bread, wheat 50 Rice casserole
Blueberries 50 Turkey ham 50 Egg white powder3
Bread, wheat 25 Lettuce 20 White rice5 120
Mayonnaise 15 Frozen carrots 30
Salad Celery 35
Red cabbage 10 Black beans, canned 48
Carrot 30 Frozen spinach 30
Celery 40 Margarine 10
Italian dressing 20 Seasonings6
Cheddar cheese 20
Day 3 Cranberry juice 180 Peach nectar 180 Milk, 2% milkfat 250 Margarine 30
Milk, 2% milkfat 180 Sandwich White rice5 120 Vanilla wafers 20
Oatmeal, cooked 150 Bread, wheat 50 Turkey breast 50
Raisins 20 Tuna 40 Turkey gravy 65
Bread, wheat 25 Egg white powder3 Green beans, canned 100
Lettuce 20 Apricots, canned 100
Mayonnaise 20 Angel food cake7 50
Dill pickle 10
Three bean salad
Kidney beans, canned 30
Green beans, canned 40
Garbanzos, canned 10
Onion flakes 1
Vinegar 10
Corn oil 5

1 Intake of the following items was adjusted to maintain individual’s energy needs: sugar, margarine, soft drinks, hard candy, jam (strawberry fruit
spread, raspberry preserves) and vanilla cookies.
2 Reconstituted from frozen concentrate.
3 Standard egg white powder (Oskaloosa Food Products, Oskaloosa, IA) used in variable amounts based on the subjects’ protein needs.
4 Spike seasoning (Modern Product, Milwaukee, WI).
5 Cooked instant rice.
6 Onion flakes 1.5 g, salt 1.5 g, black pepper 0.4 g, Beau monde seasoning 0.3 g (Specialty Brands, Burns Philip Food, San Francisco, CA).
7 Angel food cake prepared from mix (Betty Crocker, Minneapolis, MN) according to manufacturer’s directions.

mination of lymphocyte PLP concentration [preliminary results re-
ported elsewhere (17)].
Serum alkaline phosphatase activity was determined by Patholo-
gists’ Regional Laboratory (Lewiston, ID) as described by Bowers and
McComb (18) on samples taken at screening. Vitamin B-6 metabo-
lites [i.e., PLP, pyridoxamine phosphate (PMP), PL, PN and 4-PA] in
plasma and erythrocytes were determined by HPLC with fluorometric
detection (19). Recoveries of added vitamers from plasma were 88%
and 94% for PLP and PL, respectively. Recoveries of added vitamers
from erythrocytes were 64 ⫾ 9, 114 ⫾ 14, 97 ⫾ 24 and 93 ⫾ 11% for
PLP, PMP, PL and PN, respectively. Erythrocyte PLP values were
corrected for recovery. Erythrocyte alanine and aspartate aminotrans-
ferase activities (EALT and EAST) were measured with and without
added PLP (20); EALT was assayed the same day blood was drawn,
and EAST was assayed the next day on cells frozen at ⫺40°C. The
EALT and EAST activity coefficients were calculated as the ratio of
stimulated (PLP added) to unstimulated (no PLP added) activities.
Urinary, plasma and erythrocyte vitamin B-6 metabolite and
aminotransferase activity measurements were carried out under yel-
low light to prevent photodecomposition. All analyses were per-
formed in duplicate.
Statistical analyses. Data were analyzed using SAS and JMP
statistical analysis computer programs (SAS Institute, Cary, NC).
Group means and standard deviations were calculated at each time
point for all measurements. The last time point in each experimental
period was used in repeated-measures ANOVA. If 24-h urine collec-
tions were judged complete on the basis of creatinine excretion,
urinary 4-PA and total vitamin B-6 excretion were averaged over the
1780 HANSEN ET AL.

TABLE 4 RESULTS
Average daily nutrient composition of the basal diet On d 29, the plasma PLP concentration and urinary 4-PA
consumed by seven women during the adjustment and excretion of one subject were found to be 28- and 8-fold
experimental periods1 higher, respectively, than the mean values of the other seven
subjects. Therefore, we concluded that this subject did not
Energy, kJ (kcal) 6460 ⫾ 33 (1544 ⫾ 8)2 adhere to the study protocol, and her data were eliminated
Protein, g (% of energy) 56 ⫾ 3 (14) from the statistical analyses. No differences were found at any
Carbohydrate, g (% of energy) 215 ⫾ 38 (55) time point in means of urinary excretion of creatinine or
Fat, g (% of energy) 55 ⫾ 17 (31)
Vitamin B-6, mg 1.00 ⫾ 0.01 hematologic measures (data not reported). Two subjects were
Riboflavin, mg 1.6 ⫾ 0.2 given an additional iron supplement (27 mg/d elemental iron,
Niacin, mg 11.7 ⫾ 4.1 Fergon; Bayer Corporation, Morristown, NJ) when their he-
Vitamin B-12, ␮g 2.7 ⫾ 1.1 moglobin concentration fell below 120 g/L, one after wk 1 and
Folate, ␮g 237 ⫾ 77 the other after wk 5 of consuming the controlled diet. Mean
Vitamin C, mg 90 ⫾ 21 body weight at the end of the study was not significantly
Calcium, mg 984 ⫾ 116
Magnesium, mg 281 ⫾ 20 different from baseline.
Zinc, mg 7.1 ⫾ 0.7 Diet. Individual vitamin B-6 intake of subjects before the
Iron, mg 10.5 ⫾ 2.2 study, calculated from 3-d diet records, ranged from 0.9 to 2.1
Dietary fiber, g 18 ⫾ 2 mg/d (0.013– 0.024 mg vitamin B-6/g protein; Table 2). Food
composites from d 1, 2 and 3 of the basal diet, analyzed by
1 Computer analysis using Nutritionist IV (First DataBank, San
microbiological assay, contained 0.97 ⫾ 0.04, 1.02 ⫾ 0.04 and
Bruno, CA).
2 Mean ⫾ SD of the three menus. 0.97 ⫾ 0.03 mg vitamin B-6, respectively. PN glucoside was

Downloaded from jn.nutrition.org by guest on August 9, 2017

last 3 d of each period before performing statistical analyses, to
minimize the effect of day-to-day variation. When repeated-measures
ANOVA indicated significant differences among means, multiple
comparison tests were performed using least significant difference.
Pearson’s product-moment correlation coefficients were computed to
determine relationships among vitamin B-6 status measures and vi-
tamin B-6 intake. Student’s paired t test was used to compare mean
body weight at the beginning and end of the study. Statistical
comparisons were considered significant at P ⱕ 0.05.
Values for vitamin B-6 status indicators at the end of the adjust-
ment and three experimental periods were regressed on vitamin B-6
intake and the dietary vitamin B-6 to protein ratio (6). Intakes were
adjusted for bioavailability by converting supplemental vitamin B-6
to dietary vitamin B-6 equivalents [dietary vitamin B-6 equivalents
⫽ food vitamin B-6 ⫹ (1.27 ⫻ supplemental vitamin B-6)] (1).
Adequate (21) and baseline values of status indicators were used to
calculate estimates and 95% confidence intervals of vitamin B-6 EAR
(1) by inverse prediction (22), using the linear regression model
equations that were statistically significant. Weighted means were
determined on the basis of the inverse of the confidence interval
range. Recommended Dietary Allowances were calculated using the
formula: RDA ⫽ 1.2 ⫻ EAR (1), which assumes an EAR CV of 10%.
19% of the total vitamin B-6 intake in the basal diet for d 1,
21% for d 2 and 9% for d 3, with a mean (⫾SD) of 16 ⫾ 6%.
When vitamin B-6 supplements given during the experimental
periods are taken into consideration, PNG was 11, 8 and 6%
of the total vitamin B-6 intake during Periods 1, 2 and 3,
respectively. The amount of PNG in the typical American diet
has been estimated to be 10 –15% of the total vitamin B-6
content (23,24).
Urinary vitamin B-6 status measures. Mean urinary
4-PA and total vitamin B-6 excretion at baseline (d 1) and at
the end of each experimental period are presented in Table 5.
Urinary 4-PA excretion at baseline was ⱖ3.0 ␮mol, consid-
ered to be indicative of adequate vitamin B-6 status (21), for
all subjects except one (2.5 ␮mol/d). At the end of the
adjustment period (1.0 mg/d vitamin B-6 intake), mean uri-
nary 4-PA excretion had decreased 38% compared with base-
line, and only two subjects were excreting ⱖ3.0 ␮mol/d. Mean
urinary 4-PA excretion increased significantly with each suc-
cessive increase in vitamin B-6 intake, and was ⱖ3.0 ␮mol/d
for all seven subjects at the end of all three experimental
periods. By the end of Period 1 (1.5 mg/d vitamin B-6 intake),
mean urinary 4-PA excretion was not significantly different

TABLE 5
Urinary and plasma vitamin B-6 status measures of women at baseline and after consuming four levels of vitamin B-6 for a 7-d
adjustment period and three 14-d experimental periods1

Baseline, Adjustment, Period 1, Period 2, Period 3,

Vitamin B-6 intake, mg 1.4 ⫾ 0.6 1.0 1.5 2.1 2.7

␮mol/d

Urinary 4-PA 4.62 ⫾ 1.39c 2.87 ⫾ 0.93d 4.23 ⫾ 0.61c 6.77 ⫾ 0.80b 9.16 ⫾ 1.21a
Urinary TB6 0.72 ⫾ 0.24a,b 0.47 ⫾ 0.10b 0.57 ⫾ 0.13a,b 0.71 ⫾ 0.18a,b 0.85 ⫾ 0.24a

nmol/L

Plasma PLP 46.6 ⫾ 13.9a,b 29.7 ⫾ 7.1c 35.2 ⫾ 6.0b,c 43.7 ⫾ 7.2a,b,c 56.1 ⫾ 13.2a
Plasma PL 22.5 ⫾ 18.1 16.0 ⫾ 13.6 15.7 ⫾ 15.1 20.7 ⫾ 17.1 24.5 ⫾ 10.0
Plasma PN 6.29 ⫾ 1.57 6.49 ⫾ 1.15 7.15 ⫾ 1.24 6.86 ⫾ 1.50 6.17 ⫾ 1.25
Plasma TB6 75.3 ⫾ 21.6a,b 51.2 ⫾ 11.1b 58.1 ⫾ 18.0a,b 71.4 ⫾ 21.3a,b 86.8 ⫾ 19.7a
Plasma 4-PA 19.9 ⫾ 4.5 12.4 ⫾ 4.1 15.9 ⫾ 6.2 19.0 ⫾ 6.7 20.7 ⫾ 5.8

1 Mean ⫾ SD, n ⫽ 7. Values within a row with different superscript letters are significantly different, P ⱕ 0.05. Abbreviations: 4-PA, 4-pyridoxic acid;
TB6, total vitamin B-6; PLP, pyridoxal phosphate; PL, pyridoxal; PN, pyridoxine.
 VITAMIN B-6 REQUIREMENT AND RECOMMENDATIONS 1781

from baseline, but two subjects were excreting 29 and 35% less
than baseline. Mean urinary 4-PA excretion represented 59,
48, 55 and 58% of total vitamin B-6 intake during adjustment
and the three experimental periods, respectively.
Urinary total vitamin B-6 excretion at baseline was ⱖ0.5
␮mol/d, indicating adequate status (21) in all subjects. At the
end of the adjustment period, mean urinary total vitamin B-6
excretion had decreased 35% from baseline, and only three
subjects were excreting ⱖ0.5 ␮mol/d. Mean urinary total
vitamin B-6 excretion at the end of Period 3 (2.7 mg/d vitamin
B-6 intake) was significantly different from the end of the
adjustment period, and was ⱖ0.5 ␮mol/d for all subjects at the
end of all three experimental periods. By the end of Period 2
(2.1 mg/d vitamin B-6 intake), mean urinary total vitamin B-6
had returned to baseline, but two subjects were excreting 29
and 37% less than their baseline value. Mean urinary total
vitamin B-6 excretion represented 8.0, 6.4, 5.7 and 5.3% of
total vitamin B-6 intake during the adjustment and three
experimental periods, respectively.
Plasma vitamin B-6 status measures. Plasma PLP con-
centrations (Table 5) at baseline for all subjects except one
(27.6 nmol/L) were ⱖ30 nmol/L, indicating adequate vitamin
There were no significant differences in mean plasma PL,
PN or 4-PA concentrations among the periods. Mean plasma
total vitamin B-6 (i.e., PLP ⫹ PL ⫹ PN) concentrations were
significantly different at the end of the adjustment and three
experimental periods. At the end of the adjustment period,
mean plasma total vitamin B-6 was significantly lower (32%)
than baseline, reflecting the observed changes in plasma PLP
concentration. Six of seven subjects’ plasma total vitamin B-6
reached baseline by the end of Period 3.
Erythrocyte vitamin B-6 status measures. Erythrocyte
vitamin B-6 metabolite concentrations and aminotransferase
activities are given in Table 6. Erythrocyte PLP concentration
increased significantly (25%) from the adjustment period and
surpassed baseline by the end of Period 2; two of seven subjects
remained below their baseline concentration. At the end of
Period 3, mean erythrocyte PLP was significantly greater
(26%) than at the end of Period 2, and all subjects reached or
exceeded baseline. Erythrocyte PMP concentration was signif-
icantly increased from baseline (29%) and adjustment period
(18%) at the end of Period 3. Erythrocyte PL concentration
was significantly increased (39%) from the adjustment period

Downloaded from jn.nutrition.org by guest on August 9, 2017

B-6 status (21). Mean plasma PLP concentration decreased
36% and was significantly different from baseline at the end of
the adjustment period; three of seven subjects had plasma PLP
concentrations ⬍30 nmol/L. After 2 wk of consuming 1.5 mg
vitamin B-6/d, plasma PLP concentrations remained 29 – 60%
lower than baseline in four subjects, but ⬍30 nmol/L in only
one subject. At the end of Period 2 (2.1 mg/d vitamin B-6
intake), plasma PLP concentrations for four subjects were
21–35% less than baseline, but all subjects had concentrations
ⱖ30 nmol/L. By the end of Period 3 (2.7 mg/d vitamin B-6
intake), mean plasma PLP concentration had increased 89%
and was significantly different from the adjustment period.
Although mean plasma PLP concentration at the end of
Period 3 was greater than the mean baseline concentration,
three subjects had not achieved baseline concentrations of
plasma PLP. No subjects had plasma PLP concentrations ⬍20
nmol/L during any of the experimental periods.
at the end of Period 1, but did not differ significantly from
baseline during any of the periods. Erythrocyte PN concentra-
tion was not significantly different at any time point during the
study. Erythrocyte total vitamin B-6 (i.e., PLP ⫹ PMP ⫹ PL
⫹ PN) concentrations were never significantly different from
baseline, but were significantly higher than the adjustment
concentration at the end of Periods 1 (16%) and 3 (19%).
Erythrocyte alanine and aspartate activity coefficients did not
vary significantly throughout the study. Basal activity for
EALT was significantly increased at the end of Period 3
compared with all the other periods.
Alkaline phosphatase activity has been found to have an
effect on plasma PLP concentrations (25). Serum alkaline
phosphatase activity in our subjects was within the normal
range. No significant correlations were found between alkaline
phosphatase activity and plasma or erythrocyte PLP concen-
trations at baseline (d 1) or the end of the study (d 49).

TABLE 6
Erythrocyte vitamin B-6 status measures of women at baseline and after consuming four levels of vitamin B-6 for a 7-d
adjustment period and three 14-d experimental periods1

Baseline, Adjustment, Period 1, Period 2, Period 3,

Vitamin B-6 intake, mg 1.4 ⫾ 0.6 1.0 1.5 2.1 2.7

nmol/L

Erythrocyte PLP 35.5 ⫾ 6.6b,c 32.1 ⫾ 4.1c 31.5 ⫾ 11.9c 40.1 ⫾ 6.6b 50.5 ⫾ 6.6a
Erythrocyte PMP 34.5 ⫾ 4.4c 37.7 ⫾ 6.1b,c 39.8 ⫾ 7.9a,b,c 41.2 ⫾ 4.8a,b 44.6 ⫾ 6.9a
Erythrocyte PL 63.3 ⫾ 33.0a,b 48.3 ⫾ 18.9b 67.3 ⫾ 25.9a 52.9 ⫾ 16.4a,b 49.4 ⫾ 21.8a,b
Erythrocyte PN 17.5 ⫾ 7.4 15.7 ⫾ 7.0 17.5 ⫾ 4.5 15.8 ⫾ 3.1 15.9 ⫾ 5.5
Erythrocyte TB6 151 ⫾ 42a,b 134 ⫾ 26b 156 ⫾ 24a 150 ⫾ 17a,b 160 ⫾ 27a

␮kat/L RBC

EALT basal activity 0.72 ⫾ 0.27b 0.84 ⫾ 0.22b 0.74 ⫾ 0.17b 0.81 ⫾ 0.30b 1.09 ⫾ 0.40a
EAST basal activity 13.4 ⫾ 2.9c 21.4 ⫾ 3.5a,b 23.6 ⫾ 2.7a 19.8 ⫾ 4.0b 19.9 ⫾ 3.4a,b

ratio stimulated:unstimulated activity

EALT activity coefficient 1.20 ⫾ 0.12 1.18 ⫾ 0.06 1.17 ⫾ 0.06 1.16 ⫾ 0.06 1.15 ⫾ 0.04
EAST activity coefficient 1.74 ⫾ 0.25a 1.55 ⫾ 0.08b 1.58 ⫾ 0.11b 1.58 ⫾ 0.15b 1.57 ⫾ 0.14b

1 Mean ⫾ SD. Values within a row with different superscript letters are significantly different, P ⱕ 0.05. Abbreviations: PLP, pyridoxal phosphate;
PMP, pyridoxamine phosphate; PL, pyridoxal; PN, pyridoxine; RBC, red blood cells; TB6, total vitamin B-6; EALT, erythrocyte alanine aminotrans-
ferase; EAST, erythrocyte aspartate aminotransferase.
1782 HANSEN ET AL.

TABLE 7
Correlations (r) among vitamin B-6 status indicators of women consuming four levels of vitamin B-6 for a 7-d adjustment period
and three 14-d experimental periods1

Urinary Urinary Plasma Plasma Plasma Plasma Plasma Eryth. Eryth. Eryth. Eryth. Eryth. EALT EALT EAST EAST
4-PA TB6 PLP PL PN TB6 4-PA PLP PMP PL PN TB6 basal AC basal AC

B-6 intake 0.940 0.673 0.765 0.255 ⫺0.136 0.647 0.498 0.686 0.382⫺0.074⫺0.029 0.330 0.303 ⫺0.205 ⴚ0.393 0.039
B-6:pro 0.922 0.722 0.856 0.272 ⫺0.117 0.710 0.500 0.707 0.370⫺0.164⫺0.062 0.254 0.327 ⫺0.199 ⫺0.319 ⫺0.023
Urinary 4-PA 0.637 0.766 0.246 ⫺0.135 0.641 0.509 0.732 0.433 0.004 0.055 0.451 0.475 ⫺0.142 ⫺0.350 ⫺0.103
Urinary TB6 0.828 0.599 ⫺0.064 0.896 0.043 0.454 0.314⫺0.160 0.052 0.155 0.338 ⴚ0.431 ⫺0.066 ⫺0.071
Plasma PLP 0.287 ⫺0.128 0.822 0.310 0.628 0.379⫺0.288⫺0.014 0.126 0.307 ⫺0.242 ⫺0.148 0.002
Plasma PL 0.369 0.860 ⫺0.098 0.313 0.117⫺0.036 0.119 0.163 0.129 ⫺0.220 ⫺0.007 ⫺0.285
Plasma PN 0.204 0.241 0.093⫺0.105⫺0.152 0.267⫺0.064⫺0.335 0.168 ⫺0.014 ⫺0.003
Plasma TB6 0.152 0.597 0.302⫺0.212 0.093 0.177 0.254 ⫺0.174 ⫺0.087 ⫺0.174
Plasma 4-PA 0.665 0.132⫺0.224 0.054 0.142 0.127 0.294 ⫺0.306 ⫺0.269
Eryth. PLP 0.222⫺0.188 0.138 0.363 0.429 ⫺0.021 ⫺0.149 ⫺0.283
Eryth. PMP ⫺0.102 0.282 0.337 0.505 ⫺0.034 ⫺0.033 ⫺0.158
Eryth. PL 0.100 0.771 0.073 ⫺0.162 0.201 ⫺0.195
Eryth. PN 0.428 0.128 0.043 0.246 ⫺0.219
Eryth. TB6 0.415 ⫺0.150 0.329 ⴚ0.382
EALT basal ⫺0.103 0.151 ⴚ0.581
EALT AC ⫺0.147 ⫺0.072
EAST basal ⴚ0.424

Downloaded from jn.nutrition.org by guest on August 9, 2017

1 Based on n ⫽ 28 observations, except plasma pyridoxine (PN; n ⫽ 27). Significant (P ⱕ 0.05) correlations are in bold print. Abbreviations:
B-6:pro, dietary vitamin B-6 to protein ratio (mg/g); 4-PA, 4-pyridoxic acid; TB6, total vitamin B-6; PLP, pyridoxal phosphate; PL, pyridoxal; Eryth,
erythrocyte; PMP, pyridoxamine phosphate; EALT, erythrocyte alanine aminotransferase; AC, activity coefficient; EAST, erythrocyte aspartate
aminotransferase.

Correlations among vitamin B-6 status indicators. Cor-
relations among vitamin B-6 intake, the dietary vitamin B-6 to
protein ratio and vitamin B-6 status indicators are listed in
Table 7. Vitamin B-6 intake and the dietary vitamin B-6 to
protein ratio were significantly correlated with urinary 4-PA
and total vitamin B-6, plasma PLP, total vitamin B-6 and
4-PA, and erythrocyte PLP. In addition, vitamin B-6 intake
was significantly correlated with erythrocyte PMP and EAST
basal activity.
Urinary 4-PA excretion was also significantly correlated
with urinary total vitamin B-6, plasma PLP, total vitamin B-6
and 4-PA, erythrocyte PLP, PMP, total vitamin B-6 and
EAST basal activity. In addition, plasma PLP was significantly
correlated with urinary total vitamin B-6, plasma total vitamin
B-6, erythrocyte PLP and PMP. Plasma PL was significantly
correlated with urinary total vitamin B-6 and plasma total
vitamin B-6. Erythrocyte PLP was significantly correlated with
urinary total vitamin B-6, plasma total vitamin B-6 and 4-PA.
Erythrocyte PMP was also correlated with EALT basal activ-
ity. Erythrocyte total vitamin B-6 was significantly correlated
with erythrocyte PL, PN, EALT basal activity and EAST
activity coefficient.
Vitamin B-6 requirement. The EAR of vitamin B-6 was
calculated by inverse prediction, using both adequate and
baseline values (Table 8). Linear regression analyses of plasma
PLP vs. vitamin B-6 intake (adjusted for bioavailability) and
the dietary vitamin B-6 to protein ratio (adjusted for bioavail-
ability) are depicted in Figure 1A and B, respectively. Similar
analyses were performed for each of the vitamin B-6 status
indicators listed in Table 8. Regression analysis equations for
the other status indicators (y) vs. vitamin B-6 intake (x) and
the dietary vitamin B-6 to protein ratio (x), respectively, are as
follows: for urinary 4-PA, y ⫽ 2.967x ⫹ ⫺0.2680 and y
⫽ 197.5x ⫹ 0.05004; for urinary total vitamin B-6, y

TABLE 8
Vitamin B-6 requirement calculated by inverse prediction based on adequate and baseline values of vitamin B-6 status indicators
of women consuming four levels of vitamin B-6 for a 7-d adjustment period and three 14-d experimental periods1

Vitamin B-6 intake Vitamin B-6 intake Dietary B-6 to protein Dietary B-6 to protein
predicted by predicted by ratio predicted by ratio predicted by
Vitamin B-6 status indicator adequate value baseline value adequate value baseline value

mg/d mg/g

Urinary 4-PA 1.1 (0.9–1.3) 1.6 (1.3–2.0) 0.015 (0.011–0.018) 0.023 (0.017–0.029)
Urinary total vitamin B-6 1.2 (0.4–1.6) 2.4 (1.7–2.8) 0.018 (0.008–0.023) 0.034 (0.025–0.040)
Plasma PLP 1.1 (0.5–1.4) 2.5 (1.5–3.2) 0.017 (0.012–0.020) 0.035 (0.023–0.046)
Erythrocyte PLP NE 1.7 (0.9–2.3) NE 0.024 (0.012–0.032)
Weighted mean2 1.1 (0.7–1.4) 2.0 (1.4–2.5) 0.016 (0.011–0.020) 0.028 (0.019–0.038)

1 Values in parentheses are 95% confidence intervals. Abbreviations: 4-PA, 4-pyridoxic acid; PLP, pyridoxal phosphate; NE, adequate values not
established.
2 Weighted using inverse of confidence interval range as weight.
 VITAMIN B-6 REQUIREMENT AND RECOMMENDATIONS
⫽ 0.1789x ⫹ 0.2847 and y ⫽ 13.02x ⫹ 0.2719; for erythrocyte
PLP, y ⫽ 8.950x ⫹ 20.36 and y ⫽ 626.0x ⫹ 20.44. The
weighted mean of the predictions yielded an EAR and RDA of
1.1 mg/d (0.016 mg/g protein) and 1.3 mg/d (0.018mg/g pro-
tein) vitamin B-6, respectively, using adequate values (21),
and, 2.0 mg/d (0.028 mg/g protein) and 2.4 mg/d (0.031 mg/g
protein) vitamin B-6, respectively, using baseline values.
DISCUSSION
The RDA of vitamin B-6 for women ages 19 –50 y was
recently reduced from 1.6 mg/d (11) to 1.3 mg/d (1). The
Committee on the Scientific Evaluation of Dietary Reference
Intakes calculated the RDA on the basis of an EAR of 1.1
mg/d, the intake of vitamin B-6 required for a plasma PLP
concentration of 20 nmol/L. The Committee determined this
adequate value for plasma PLP concentration from a single
study (9) in which plasma PLP concentrations of 94 healthy
men ages 18 – 68 y, consuming self-selected diets, were as-
sessed. The lowest value for plasma PLP concentration found
in this group (5 ng/mL or 20 nmol/L) was arbitrarily consid-
ered as the cut-off value for adequate status.
FIGURE 1 Linear regression analysis of plasma pyridoxal phos-
phate (PLP) concentration vs. vitamin B-6 intake (A) and the dietary
vitamin B-6 to protein ratio (B) for 7 women consuming a controlled diet
with four levels of vitamin B-6. Intakes were adjusted by converting
supplemental vitamin B-6 to dietary vitamin B-6 equivalents [dietary
vitamin B-6 equivalents ⫽ food vitamin B-6 ⫹ (1.27 ⫻ supplemental
vitamin B-6)] (1). Inverse regression calculations using adequate (21)
and baseline values of PLP concentration predict Estimated Average
Requirements for vitamin B-6 of 1.1 and 2.5 mg/d, respectively, and
dietary vitamin B-6 to protein ratios of 0.017 and 0.035 mg/g, respec-
tively.
1783
Sauberlich (26) proposed that interpretation of biochemi-
cal results in nutritional status assessment use a statistical
approach in classifying the following three states of risk: 1)
high, 2) borderline or moderate, and 3) low. Subjects with
values below the 2.5th percentile would be classified as high
risk, those between the 2.5th and 30th percentile would be
considered borderline (marginally or subclinically deficient),
and those above the 30th percentile would be considered to
have acceptable values or adequate status. Applying Sauberli-
ch’s approach to baseline data obtained in our laboratories
(3–5,27–29) from 60 healthy, unsupplemented women ages
19 –50 y consuming self-selected diets with a plasma PLP
concentration (mean ⫾ SD) of 42.8 ⫾ 19.1 nmol/L (range:
14.2–109 nmol/L) and urinary 4-PA excretion of 5.25 ⫾ 2.59
␮mol/d (range: 2.24 –20.22 ␮mol/d), we determined an ac-
ceptable value for plasma PLP concentration of 31.1 nmol/L,
indicating adequate status, and 4.07 ␮mol/d for urinary 4-PA
excretion. Thus, when calculating an EAR for vitamin B-6 we
used the adequate values published by Leklem (21), e.g., 30
nmol/L for plasma PLP, rather than the 20 nmol/L plasma PLP
concentration used by the DRI Committee.
Many of the studies that have assessed vitamin B-6 require-
Downloaded from jn.nutrition.org by guest on August 9, 2017
ments have determined the intake that returns status indica-
tors to their prestudy baseline values (6,7,30,31). This ap-
proach has been criticized because the subjects have been
motivated healthy individuals consuming self-selected diets or
diets that contained 1.5–2.0 mg vitamin B-6, and the assessed
requirements generally are similar to or higher than the base-
line vitamin B-6 intake (1). The subjects in the study reported
here had a baseline intake of 1.4 ⫾ 0.6 mg/d (0.020 ⫾ 0.007
mg/g protein) and the assessed requirement predicted by base-
line status indicators was 2.0 mg/d (0.028 mg/g protein), which
is higher than the predicted requirement based on adequate
values. Therefore, following the guidelines suggested by the
DRI committee, we base our recommendations on inverse
predictions using adequate values rather than baseline values.
In the 1989 Recommended Dietary Allowances (11), the
RDA for vitamin B-6 was based on protein intake (i.e., 0.016
mg/g protein). Increased protein intake has been shown to
decrease several measures of vitamin B-6 status (3,32–35).
Although Pannemans et al. (36) failed to show an effect of
dietary protein on vitamin B-6 status indicators in elderly
subjects, young subjects consuming a high protein diet ex-
creted less urinary 4-PA compared with subjects consuming a
low protein diet. Another study (31) reported that higher
vitamin B-6 intakes were necessary to normalize plasma PLP
concentrations in elderly subjects consuming a high protein
diet vs. subjects consuming a low protein diet. In the study
reported here, because protein intake was based on body
weight, we could separate subjects into two body weight/
protein intake groups. The lower body weight/protein group (n
⫽ 4) had a mean (⫾SD) body weight of 54 ⫾ 2 kg and protein
intake of 65 ⫾ 3 g/d, whereas the higher body weight group (n
⫽ 3), had values of 66 ⫾ 1 kg and 79 ⫾ 2 g/d, respectively. At
baseline, the higher body weight/protein group had higher
plasma PLP concentrations (52.5 ⫾ 8.3 nmol/L vs. 42.2
⫾ 16.6 nmol/L for the lower body weight/protein group). By
the end of Period 3, the higher body weight/protein group had
a plasma PLP concentration of 43.0 ⫾ 1.9 nmol/L compared
with 65.9 ⫾ 6.9 nmol/L for the lower body weight/protein
group. Thus, a 14 g higher protein intake resulted in ⬃23
nmol/L lower plasma PLP concentration. There was no similar
effect on other measures of vitamin B-6 status.
Furthermore, if plasma PLP concentrations of the body
weight/protein groups are regressed separately on adjusted vi-
tamin B-6 intake, the inverse prediction of EAR for the higher
1784 HANSEN ET AL.

group is 1.3 mg/d compared with 1.0 mg/d for the lower group.
When plasma PLP concentrations are regressed on the dietary
vitamin B-6 to protein ratio, however, the predicted EAR is
similar for both groups (0.016 and 0.017 mg/g for the lower
and higher groups, respectively). These data suggest a signifi-
cant effect of protein intake on plasma PLP concentration and
the importance of considering dietary protein when establish-
ing EAR and RDA for vitamin B-6. However, because the
effects of increased body weight and increased protein intake
cannot be separated in the current study, the possibility that
increased body weight increases the vitamin B-6 requirement
is an alternative explanation of this effect.
When we compared the fit of regression lines relating
vitamin B-6 status measures to either adjusted vitamin B-6
intake or the dietary vitamin B-6 to protein ratio, for all
measures except urinary 4-PA excretion, the r value was
higher using the dietary vitamin B-6 to protein ratio. We then
combined data from the current study and four other recent
studies involving young subjects (3,5–7) to investigate
whether plasma PLP concentration (Fig. 2A and B) and
urinary 4-PA excretion (Fig. 3A and B) were more highly
correlated with vitamin B-6 intake or the dietary vitamin B-6

Downloaded from jn.nutrition.org by guest on August 9, 2017

to protein ratio, and to calculate an EAR and RDA based on
the combined data using the values for adequate status sug-
gested by Leklem (21). Intakes were adjusted by converting
supplemental vitamin B-6 to dietary vitamin B-6 equivalents.
Means of plasma PLP concentration and urinary 4-PA excre-
tion were weighted, using the number of subjects (n) as
weights (total sum weights ⫽ 177 observations). Regression
analysis of urinary 4-PA on vitamin B-6 intake and the dietary
vitamin B-6 to protein ratio produced a better fit when urinary
4-PA data were transformed by taking the square root. For the
combined data, regression on adjusted vitamin B-6 intake
resulted in a better fit (higher r values) than regression on the
dietary vitamin B-6 to protein ratio. The EAR of vitamin B-6
was determined to be 1.2 mg/d (0.015 mg/g protein) and 1.3
mg/d (0.015 mg/g protein) by inverse prediction using ade-
quate values of plasma PLP and urinary 4-PA (21), respec-
tively. The RDA, assuming a CV in vitamin B-6 requirement
of 10%, was calculated to be 1.4 –1.6 mg/d (0.018 mg/g pro-
tein). The predicted EAR and RDA for vitamin B-6 from the
current study (1.1 mg/d or 0.016 mg/g protein and 1.3 mg/d or
0.018 mg/g protein) agree well with the predicted EAR using
the combined data.
Because the RDA is calculated on the basis of variability of
the EAR, the following question remains to be answered: what
is the variability of the EAR? The DRI Committee assumes a
CV of 10% in the absence of evidence of variability of the
EAR. In a previous study, the average CV in the vitamin B-6
requirement based on six status measures was estimated to be
28% (6). In the current study, the CV in requirement based on
plasma PLP was 20% for vitamin B-6 intake and 13% for the
dietary vitamin B-6 to protein ratio. These data suggest that
the RDA will be underestimated by assuming a CV of 10%.
Using the EAR prediction of 1.1 mg/d and a CV of 20%, the
RDA is calculated to be 1.5 mg/d. Similarly, using the EAR
prediction for the vitamin B-6 to protein ratio of 0.016 mg/g
and a CV of 13%, the RDA is calculated to be 0.020 mg/g
protein. Assuming a CV of 20% for vitamin B-6 intake and
13% for the dietary vitamin B-6 to protein ratio, the combined
studies data suggest an RDA of 1.7 mg/d and 0.018 mg/g
protein, respectively.
Another question to be considered is whether plasma PLP
concentration is the most appropriate measurement to use as a
standard for setting requirements. During abnormal conditions
[e.g., acute phase of myocardial infarction (37)] or treatment
FIGURE 2 Linear regression analysis of plasma pyridoxal phos-
phate (PLP) concentration vs. vitamin B-6 intake (A) and the dietary
vitamin B-6 to protein ratio (B) from the current study and four other
recent studies (3,5–7) of women consuming controlled diets with known
amounts of vitamin B-6 and protein. Intakes were adjusted by convert-
ing supplemental vitamin B-6 to dietary vitamin B-6 equivalents [dietary
vitamin B-6 equivalents ⫽ food vitamin B-6 ⫹ (1.27 ⫻ supplemental
vitamin B-6)] (1). Inverse regression calculations using adequate value
(21) of PLP concentration predict an Estimated Average Requirement
for vitamin B-6 of 1.2 mg/d and a dietary vitamin B-6 to protein ratio of
0.015 mg/g.
with drugs [e.g., theophylline (38)], and normal life-cycle
stages such as pregnancy (39), plasma PLP concentration may
not be indicative of vitamin B-6 status (40). For the purpose of
setting requirements for healthy nonpregnant people, how-
ever, using plasma PLP is appropriate because this status mea-
sure correlates significantly with vitamin B-6 intake and the
dietary vitamin B-6 to protein ratio (Table 7).
In the present study, urinary 4-PA excretion and erythro-
cyte PLP concentration were also strongly correlated with
both vitamin B-6 intake and the dietary vitamin B-6 to pro-
tein ratio. The DRI Committee rejected using urinary 4-PA
excretion for status assessment because 4-PA excretion re-
sponds rapidly to changes in dietary intake (8), thus reflecting
only recent intake (1). Erythrocyte PLP has been suggested as
a more relevant measure because the site of PLP coenzyme
function is intracellular (40). However, few studies assessing
vitamin B-6 requirements have measured erythrocyte metab-
 VITAMIN B-6 REQUIREMENT AND RECOMMENDATIONS
FIGURE 3 Linear regression analysis of the square root (sqrt) of
urinary 4-pyridoxic acid (4-PA) excretion vs. vitamin B-6 intake (A) and
the dietary vitamin B-6 to protein ratio (B) from the current study and
four other recent studies (3,5–7) of women consuming controlled diets
with known amounts of vitamin B-6 and protein. Intakes were adjusted
by converting supplemental vitamin B-6 to dietary vitamin B-6 equiva-
lents [dietary vitamin B-6 equivalents ⫽ food vitamin B-6 ⫹ (1.27
⫻ supplemental vitamin B-6)] (1). Inverse regression calculations using
the sqrt of urinary 4-PA adequate value (21) predict an Estimated
Average Requirement for vitamin B-6 of 1.3 mg/d and a dietary vitamin
B-6 to protein ratio of 0.015 mg/g.
olites (5,6); consequently, adequate values for erythrocyte PLP
concentration have not been established.
Erythrocyte aminotransferase activation by PLP (i.e.,
EALT and EAST activity coefficient or percentage stimula-
tion) is useful as a long-term measure of vitamin B-6 status
(21), but its use in assessing requirements is limited in the
present study because of the short length of the experimental
periods. Although EAST basal activity showed a significant
correlation with vitamin B-6 intake, it was paradoxically neg-
ative. One possible explanation is a lag time between changes
in intake and changes in erythrocyte enzyme activity.
Plasma PL has also been suggested as an indicator of vita-
min B-6 status (21). In agreement with a previous study
conducted in our laboratory (6), plasma PL concentration was
not significantly correlated with vitamin B-6 intake or the
dietary vitamin B-6 to protein ratio. Therefore, it appears that
1785
plasma PL has limited usefulness as a measure of vitamin B-6
status.
Ideally, the best status measure to use when determining
vitamin B-6 adequacy would be a functional measure related to
a specific health outcome. In epidemiologic studies, low vita-
min B-6 intake has been associated with increased risk of heart
disease (41– 43) and cancer (44 – 47). Future research may
further elucidate the mechanisms for these associations and
help define a functional measure upon which recommenda-
tions for vitamin B-6 intake can be based. Until such a
measure is determined, selecting adequate cut-off values for
vitamin B-6 status measures will remain controversial.
In conclusion, predicting the EAR and RDA on the basis of
adequate values of commonly measured status indicators cal-
culated by the method suggested by Sauberlich (26) is the best
approach available in the absence of status indicators linked to
specific health outcomes (e.g., prevention of heart disease,
cancer or other chronic diseases). The data presented here
combined with previously published data suggest an EAR of
1.1–1.2 mg/d (0.015– 0.016 mg/g protein) and an RDA of
1.5–1.7 mg/d (0.018 – 0.020 mg/g protein) for young women.
Downloaded from jn.nutrition.org by guest on August 9, 2017
ACKNOWLEDGMENTS
We gratefully acknowledge the technical assistance of Hedy Her-
rick, Jim Ridlington and Karin Hardin. We thank Marc Evans for his
statistical advice, and the subjects for their invaluable cooperation.
LITERATURE CITED
1. Institute of Medicine (1998) DRI Reference Intakes for Thiamin, Ribo-
flavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and
Choline. National Academy Press, Washington, DC.
2. Brown, R. R., Rose, D. P., Leklem, J. E., Linkswiler, H. & Anand, R.
(1975) Urinary 4-pyridoxic acid, plasma pyridoxal phosphate, and erythrocyte
aminotransferase levels in oral contraceptive users receiving controlled intakes of
vitamin B6. Am. J. Clin. Nutr. 28: 10 –19.
3. Hansen, C. M., Leklem, J. E. & Miller L. T. (1996) Vitamin B-6 status of
women with a constant intake of vitamin B-6 changes with three levels of dietary
protein. J. Nutr. 126: 1891–1901.
4. Hansen, C. M., Leklem, J. E. & Miller, L. T. (1996) Vitamin B-6 status
indicators decrease in women consuming a diet high in pyridoxine glucoside. J.
Nutr. 126: 2512–2518.
5. Hansen, C. M., Leklem, J. E. & Miller, L. T. (1997) Changes in vitamin
B-6 status indicators of women fed a constant protein diet with varying levels of
vitamin B-6. Am. J. Clin. Nutr. 66: 1379 –1387.
6. Huang, Y. C., Chen, W., Evans, M. A., Mitchell, M. E. & Shultz, T. D.
(1998) Vitamin B-6 requirement and status assessment of young women fed a
high-protein diet with various levels of vitamin B-6. Am. J. Clin. Nutr. 67: 208 –220.
7. Kretsch, M. J., Sauberlich, H. E., Skala, J. H. & Johnson, H. L. (1995)
Vitamin B-6 requirement and status assessment: young women fed a depletion
diet followed by a plant- or animal-protein diet with graded amounts of vitamin
B-6. Am. J. Clin. Nutr. 61: 1091–1101.
8. Lui, A., Lumeng, L., Aronoff, G. R. & Li, T.-K. (1985) Relationship
between body store of vitamin B6 and plasma pyridoxal-P clearance: metabolic
balance studies in humans. J. Clin. Lab. Med. 106: 491– 497.
9. Lumeng, L. & Li, T-K. (1974) Vitamin B6 metabolism in chronic alcohol
abuse. Pyridoxal phosphate levels in plasma and the effects of acetaldehyde on
pyridoxal phosphate synthesis and degradation in human erythrocytes. J. Clin.
Invest. 53: 693–704.
10. Harris, A. L. (1969) Determination of D-xylose in urine for the D-xylose
absorption test. Clin. Chem. 15: 65–71.
11. National Research Council (1989) Recommended Dietary Allowances,
10th ed. National Academy Press, Washington, DC.
12. Horwitz, G.W. (1980) Official Methods of Analysis of the AOAC, 13th
ed., p. 768. AOAC, Washington, DC.
13. Kabir, H., Leklem, J. E. & Miller, L. T. (1983) Measurement of glycos-
ylated vitamin B-6 in food. J. Food Sci. 48: 1422–1425.
14. Pino, S., Bennotte, J. & Gordnya, H. (1965) An automated method for
urine creatinine which does not require a dialyzer module. Clin. Chem. 11:
664 – 666.
15. Gregory, J. F., III & Kirk, J. R. (1979) Determination of urinary 4-pyri-
doxic acid using high performance liquid chromatography. Am. J. Clin. Nutr 32:
879 – 883.
16. Miller, L. T. & Edwards, M. (1981) Microbiological assay of vitamin B-6
in blood and urine. In: Methods in Vitamin B-6 Nutrition (Leklem, J. E. & Reynolds,
R. D., eds.), pp. 45–55. Plenum Press, New York, NY.
17. Kwak, H. K., Hansen, C., Hardin, K., Ridlington, J., Leklem, J. E. & Shultz,
1786 HANSEN ET AL.
T. D. (1999) A positive effect of vitamin B-6 on the immune response of young
women. FASEB J. 13: A699 (abs.).
18. Bowers, G. N., Jr. & McComb, R. B. (1975) Measurement of total
alkaline phosphatase activity in human serum. Clin. Chem. 21: 1988 –1995.
19. Sharma, S. K. & Dakshinamurti, K. (1992) Determination of vitamin B6
vitamers and pyridoxic acid in biological samples. J. Chromatogr. 578: 45–51.
20. Woodring, M. J. & Storvick, C. A. (1970) Effect of pyridoxine supple-
mentation on glutamic-pyruvic transaminase and in vitro stimulation in erythro-
cytes of normal women. Am. J. Clin. Nutr. 23: 1385–1395.
21. Leklem, J. E. (1990) Vitamin B-6: a status report. J. Nutr. 120: 1503–
1507.
22. Ott, R.L. (1993) An Introduction to Statistical Methods and Data Anal-
ysis, 4th ed. Wadsworth Publishing Company, Belmont, CA.
23. Andon, M. B., Reynolds, R. D., Moser-Veillon, P. B. & Howard, M. P.
(1989) Dietary intake of total and glycosylated vitamin B-6 and vitamin B-6
nutritional status of unsupplemented lactating women and their infants. Am. J.
Clin. Nutr. 50: 1050 –1058.
24. Gregory, J. F., Trumbo, P. R., Bailey, L. B., Toth, J. P., Baumgartner, T. B.
& Cerda, J. J. (1991) Bioavailability of pyridoxine-5-␤-D-glucoside determined
in humans by stable isotope methods. J. Nutr. 121: 177–186.
25. Whyte, M. P., Mahuren, J. D., Vrabel, L. A. & Coburn, S. P. (1985)
Markedly increased circulating pyridoxal 5⬘-phosphate levels in hypophosphata-
sia: alkaline phosphatase acts in vitamin B-6 metabolism. J. Clin. Invest. 76:
752–756.
26. Sauberlich, H. E. (1999) Introduction. In: Laboratory Tests for the
Assessment of Nutritional Status, 2nd ed., pp. 3– 8. CRC Press, Boca Raton, FL.
27. Hansen, C. M. (1995) Effect of Vitamin B-6 Intake, Protein Intake and
Bioavailability on Vitamin B-6 Status of Women. Doctoral thesis, Oregon State
University, Corvallis, OR.
28. Shultz, T. D. (1980) Comparative Nutrient Intake and Biochemical
Interrelationships among Healthy Vegetarian and Nonvegetarian Seventh-Day
Adventists, Nonvegetarians, and Hormone Dependent Cancer Subjects. Doctoral
thesis, Oregon State University, Corvallis, OR.
29. Shultz, T. D. & Leklem, J. E. (1981) Urinary 4-pyridoxic acid, urinary
vitamin B6, and plasma pyridoxal phosphate as measures of vitamin B6 status
and dietary intake in adults. In: Methods in Vitamin B6 Nutrition (Leklem, J. E. &
Reynolds, R. D., eds.), pp. 297–320. Plenum Press, New York, NY.
30. Meydani, S. N., Ribaya-Mercado, J. D., Russell, R. M., Sahyoun, N.,
Morrow, F. D. & Gershoff, S. N. (1991) Vitamin B-6 deficiency impairs inter-
leukin 2 production and lymphocyte proliferation in elderly adults. Am. J. Clin.
Nutr. 53: 1275–1280.
31. Ribaya-Mercado, J. D., Russell, R. M., Sahyoun, N., Morrow, F. D. &
Gershoff, S. N. (1991) Vitamin B-6 requirements of elderly men and women. J.
Nutr. 121: 1062–1074.
32. Baker, E. M., Canham, J. E., Nunes, W. T., Sauberlich, H. E. & McDowell,
M. E. (1964) Vitamin B6 requirement for adult men. Am. J. Clin. Nutr. 15:
59 – 66.
33. Canham, J. E., Baker, E. M., Harding, R. S., Sauberlich, H. E. & Plough,
I. C. (1969) Dietary protein—its relationship to vitamin B6 requirements and
function. Ann. N.Y. Acad. Sci. 166: 16 –29.
34. Miller, L. T., Leklem, J. E. & Shultz, T. D. (1985) The effect of dietary
protein on the metabolism of vitamin B-6 in humans. J. Nutr. 115: 1663–1672.
35. Miller, L. T. & Linkswiler, H. (1967) Effect of protein intake on the
development of abnormal tryptophan metabolism by men during vitamin B-6
depletion. J. Nutr. 93: 53–59.
36. Pannemans, D.L.E., van den Berg, H. & Westerterp, K. R. (1994) The
influence of protein intake on vitamin B-6 metabolism differs in young and elderly
humans. J. Nutr. 124: 1207–1214.
37. Vermaak, V.J.H., Barnard, H. C., Van Daelen, E.M.S.P., Potgieter, G. M.,
Van Jaarsveld, H. & Myburgh, S.J.S. (1988) Compartmentalization of pyri-
doxal-5⬘-phosphate during the acute phase of myocardial infarction. Klin.
Wochenschr. 66: 428 – 433.
38. Ubbink, J. B., Delport, R., Becker, P. J. & Bissport, S. (1989) Evidence
of a theophylline induced vitamin B-6 deficiency due to non-competitive inhibition
of pyridoxal kinase. J. Lab. Clin. Med. 113: 15–22.
39. Barnard, H. C., De Kock, J. J., Vermaak, W.J.H. & Potgieter, G. M.
(1987) A new perspective in the assessment of vitamin B-6 status during
pregnancy. J. Nutr. 117: 1303–1306.
40. Vermaak, V.J.H., Ubbink, J. B., Barnard, H. C., Potgieter, G. M., van
Jaarsveld, H. & Groenwald, A. J. (1990) Vitamin B-6 nutrition status and
cigarette smoking. Am. J. Clin. Nutr. 51: 1058 –1061.
41. Robinson, K., Mayer, E. L., Miller, D. P., Green, R., van Lente, F., Gupta,
A., Kottke-Marchant, K., Savon, S. R., Selhub, J., Nissen, S. E., Kutner, M., Topol,
E. J. & Jacobsen, D. W. (1995) Hyperhomocysteinemia and low pyridoxal
phosphate: common and independent reversible risk factors for coronary artery
disease. Circulation 92: 2825–2830.
42. Verhoef, P., Stampfer, M. J., Buring, J. E., Gaziano, J. M., Allen, R. H.,
Stabler, S. P., Reynolds, R. D., Kok, F. J., Hennekens, C. H. & Willett, W. C.
Downloaded from jn.nutrition.org by guest on August 9, 2017
(1996) Homocysteine metabolism and risk of myocardial infarction: relation with
vitamins B6, B12, and folate. Am. J. Epidemiol. 143: 845– 859.
43. Rimm, E. B., Willett, W. C., Hu, F. B., Sampson, L., Colditz, G. A.,
Manson, J. E., Hennekens, C. & Stampfer, M. J. (1998) Folate and vitamin B-6
from diet and supplements in relation to risk of coronary heart disease among
women. J. Am. Med. Assoc. 279: 359 –364.
44. Slattery, M. L., Potter, J. D., Coates, A., Ma, K. N., Berry, T. D., Duncan,
D. M. & Caan, B. J. (1997) Plant foods and colon cancer: an assessment of
specific foods and their related nutrients (United States). Cancer Causes Control
8: 575–590.
45. Key, T. J., Silcocks, P. B., Davey, G. K., Appleby, P. N. & Bishop, D. T.
(1997) A case-control study of diet and prostate cancer. Br. J. Cancer 76:
678 – 687.
46. Kaaks, R., Tuyns, A. J., Haelterman, M. & Riboli, E. (1998) Nutrient
intake patterns and gastric cancer risk: a case-control study in Belgium. Int. J.
Cancer 78: 415– 420.
47. Stolzenberg-Solomon, R. Z., Albanes, D., Nieto, F. J., Hartman, T., Tan-
grea, J. A., Rautalahti, M., Selhub, J., Virtamo, J. & Taylor, P. R. (1999) Pan-
creatic cancer risk and nutrition-related methyl-group availability indicators in
male smokers. J. Natl. Cancer Inst. 91: 535–541.
 Erratum

Hansen, C. M., Schultz, T. D., Kwak, H. K., Memon, H. S. and Leklem, J. E. (2001)
Assessment of Vitamin B-6 Status in Young Women Consuming a Controlled Diet
Containing Four Levels of Vitamin B-6 Provides an Estimated Average Requirement and
Recommended Dietary Allowance. J. Nutr. 131: 1777–1786.
On page 1783, figures 1A and 1B contain a concentration error in the labeling of the
y-axis. The corrected figures y-axis labels read “plasma pyridoxal phosphate (nmol/L)”
and appear below.

0022-3166/01 $3.00 © 2001 American Society for Nutritional Sciences. J. Nutr. 131: 2224, 2001.

2224