Nutrient intake and health status Chemical analyses of diets using portion sampling technique13

of vegans. the duplicate

MohammedAbdulla, M.D., IngridAndersson,4 Nils-Georg Asp, M.D., Knud Berthelsen, M. D., Dowen Birkhed, D.D. S., Ingrid Dencker,5 Claes-G#{246}ran Johansson, M. Sc. Margaretha , J#{224}gerstad, Ph. D., Kurt Kolar, M. Sc., Baboo M. Nair, Techn.Dr., Peter Nilsson-Ehle, M.D., Ake Nord#{233}n, M.D., Solvig Rassner, B. Sc. Bj#{246}rn kesson, , A M. D., and Per-A me Ockerman, M. D.

ABSTRACT vegans sampling compared originated (three with from fatty compared kcal) the kcal) kcal, 4.7 156 technique.

A men

strict

vegetarian three chemical figures with of total Dietary sucrose of similar kcal), kcal and similar fat women)

diet analyses, ofa 40% in fiber to both versus copper selenium kcal

(vegan collected the normal in VD in MD, was MD (351 types 110 normal versus but

diet copies nutrient mixed

(VD)l of

was 24-h

investigated. diets was diet. diet The using In the (MD). VD in the kcal). kcal) amount (9 versus kcal) kcal) were were and of 6.5 much amino vegan the VD

Six determined 30% Linoleic contained acids diet Among sodium potassium mg!l000 nearly in folic sg!day Serum of the were

middle-aged duplicate in detail ofthe acid 24 by (29 the (53 (56 doubled. lower acid (MD: reference low for in the the was portion and energy the

and By

composition Swedish mixed in intake 5 times versus 391 21 mg!l000 but kcal), 0.7 mg!l000 17 tg!l000 the iron MD). higher g!1000

Downloaded from www.ajcn.org by guest on May 1, 2011

corresponding fat acid compared (60% to 30 g!l000 and concentration were magnesium mg!l000 .eg!l00O

Swedish 8% the

dominant
1000 exceeded 6 g!l000 nutrients mmol,lOOO mmol!l000 (6.5 (39 versus versus kcal

g protein! vegans inorganic versus versus kcal). Iodine VD. (301 versus 3 to 4 sg! levels group. their 49 30 zinc versus

ofessential

the recommendations.

about (18 versus of diets mg!l000

calcium in (300 and

(2 versus (5 versus

selenium 90 tg/l000 day). of

In

even being undetectable kcal in MD) but the clinical as none Am. J. C/in. signs well of as the of nutritional their serum vegans

in several 24-h intake ofvitamin deficiency lipoproteins was overweight

diets. The VD was rich B12 was only 0.3 to 0.4 were were and observed near their the in blood the vegans. range pressures lower

No the addition,

protein

vegans

age.
KEY ents, vitamin

Nutr. 34: 2464-2477, Vegan diet, protein, nutrient duplicate amino status,

1981.
portion acids, blood sampling carbohydrates, parameters technique, dietary chemical fiber. minerals. analyses trace of nutrielements,

WORDS fatty acids, B12, folic

sterols, acid,

Introduction
l.A.,

From

the

Unit

for

Community

Care

Sciences

(MA.,

ID., Mi., AN.), Dalby, Sweden; the Department Chemical Centre (N-GA., C-G.J., B.M.N.), Vegans eat no food of animal origin. With ofNutrition, University of Lund, Lund, Sweden: the Department of respect to global economy, vegans are of inCariology (D.B.), University of Lund, School of Denterest, as their dietary habit may offer clues tistry. Malm#{246}, Sweden; the Swedish Meat Research Into whether a strictly vegetarian diet is nutri- stitute (K.K.), K#{228}vlinge, Sweden; and the Department of Clinical Chemistry (P.N-E., SR.,BA., P-A.O.), Univertionally adequate for maintaining health. In sity Hospital, Lund, Sweden. a few clinical studies (1-3) in Great Britain 2 Supported by a special grant within the Dalby Comand the United States, the health status of munity Care Sciences program from the Swedish Navegans has been demonstrated not to differ tional Board of Health and Social Welfare and by grants from that ofsubjects consuming a mixed diet. from The Swedish Medical Research Council (Project Foundations of Ekhaga, A. P#{225}hlson, Kiholms In fact, detailed clinical studies (4) have3967), raised the possibility that the vegan diet might h#{228}lsohem, and H#{228}lsokostr#{226}det. 3 Address reprint requests to: Margaretha J#{227}gerstad, have protective properties against certain disPh.D., Department of Nutrition, University of Lund, ease. Attention has, for instance, been drawn Chemical Centre, P.O.B. 740, 5-220 07 Lund, Sweden. to the hypocholesterolemic effect of vegetable 4 Technician. Dietitian. Nutrition 34: NOVEMBER 1981

2464

The

American

Journal

of

Clinical

1981, pp. 2464-2477.

American Society for

Printed
Clinical

in U.S.A.
Nutrition

NUTRIENT

INTAKE

AND

HEALTH

STATUS

OF

VEGANS

2465

a regimen usually followed by Swedish vegans. protein (5). Patients with severe angina pec- dietary has been established by its director (10). tons have been claimed benefit from vegan diets (6). Recently, Sanders (7) summarized Vegan diet the literature regarding the nutritive value of The vegan menu is very similar from day to day. The a vegan diet and the health status of vegans. variation consist of a wide selection of fruits, berries, Information about the nutrient quality of vegetables. roots, nuts, and pulses. All animal foods. a vegan diet is, however, usually based on including milk and dairy products. have been eliminated. pepper. coffee, and Indian tea are not used. calculations from food tables. In the present Salt, sugar. A typical menu of the vegan diet of F#{246}llingeg#{226}rdenis study, we have applied the duplicate portion summarized in Table I. technique, which is based on chemical analyses of duplicates of 24-h food collections as Food sampling described previously (8, 9). Hence, the prin- Two of us (IA and ID) visited F#{246}llingeg#{226}rden to cipal aim of the present investigation was assess to the degree of individual variation in food intake obtain more detailed data of the composition as a basis for the size of our food collection program. on our observations during this visit, we decided of dietary fats, sterols, proteins, carbohy- Based six subjects, each being studied during 4 drates, minerals, trace elements, and vitamins to include consecutive days, and thus providing a total of twentyin the vegan diet. Trace elements in vegan four, 24-h samples for chemical analysis. diets have been investigated only with respect The six subjects were studied while staying at F#{246}llingeg#{226}rden, and were therefore eating food from a comto iron (7). Although it is well known that the The participants were informed about the vegan diet is practically lacking in vitamin mon source. aim of the study and the duplicate portion sampling B12, analysis of the actual daily intake by technique. The importance ofmaintaining the usual food vegans has not been reported earlier (7). habits was stressed. The volunteer was responsible for preparing a duplicate food sample over each 24-h period, A medical examination that included exduplicated as exactly as possible by visual measurement. tensive clinical laboratory tests has been unIn addition a protocol was made up every day in which dertaken to evaluate the nutritional status of the participant recorded the type of food consumed and vegans in relation to the composition of their rough estimations of the quantities consumed expressed diet. as spoonfuls, cupfuls, etc.

Downloaded from www.ajcn.org by guest on May 1, 2011

Materials
Participants

and

methods

The sumed described

duplicate between earlier

portion, the meals, (8, 9).

including were

drinking collected

and

fluids confrozen as

Urinary samples Three healthy men and three healthy women (aged Urine samples were obtained on the same days as the 49 to 55 yr) who had followed a vegan dietary regimen food collections. The urine was collected in bottles confor at least 3 yr were studied. The six subjects were all taming 5 ml conc. HC1 to prevent bacterial contaminaselected and invited to F#{246}llingeg#{226}rdenby its director. and growth. The total daily volume was measured, F#{243}llingeg#{226}rdens a vegan i center in central Sweden where tion

TABLE A typical Breakfast

I vegan

diet

at

F#{246}llingeg#{225}rden with berries (apple, sesame seed (strawberries, pear, plum), milk and powdered dates, stewed fruit bilberries), dried fruit (dates, figs, sprouts (wheat. sesame seed), lentils, (apple, apricot, raisins, apricots, flax seed, wheat

Buckwheat porridge plum, pear), stewed plums), fresh fruit

bran. Lunch
Raw food (tomatoes, cucumber, finely minced carrots, oak leaf, fermented white cabbage), mixed salads (boiled or raw common beets, boiled or raw cucumber, parsley, dill, garlic or fermented vegetables, tomatoes, and lettuce fermented vegetables and leek), warm dishes often based on a stuffing mix.* Raw food and mixed salads and a hot together with bread, such as buckwheat is only served at dinner). between Mixture The herb of cereals, bulk teas, of the fruit, beverages is served and vegetable juices protein, soya flour, soup boiled rolls, whole from roots and vegetables. grain bread or crisp, hard mixed vegetables, potatoes. fermented or white cabbage,

Dinner

The rye

soup bread.

is served (Bread

Meals

between potatoe

meals flakes,

as garlic

water

(body powder.

temperature). and spices

several (imported

kinds from

of West

vegetable

Germany).

2466

ABDULLA

ET

AL.
U

E
a i-

0 =

i
C.

. :

.o

. cu

..

.-

(,

: E
U

<.
s

,
0.-.
.

r
.

._0

0-

0.nU U

.E
0

. :
0

?.

.
0 U D

.ra
ci. .a

n. . .

Downloaded from www.ajcn.org by guest on May 1, 2011

.0

U U

-.ac

;.

0-a .

UQ..

.9
c-Wei

ri

nD

Eu

Eaau
ci, <
S

-E 2:
0

u
I
-

<

( U

a #{149}2
, .

a
0 II.. U .C-

e
..
it,

fi
.u
.

,,
c) .-.i r.1

. _t
I

bn

#{149}

>..r

.9 0
.a
.

U> U

.a
a-

-E

.a
-

-00

.<
u
..

-.i

z
Os

z
00
0

LU

.-..

-..

-..-

...
0

00

.E
E

r1
l

o
00

0 00

sO

ri F-.

00 sO

ri

ri

:i
_

crl sO

a
if

1,5

its sO

Os r-

O.

00

:
e ifs
-

:
c

<

ir

,,

rl ifs

ifs

LL.

IL.

<. I-Li .3, >

r;
;;.

e.;s
>. >. >. >

NUTRIENT and was a representative sample ofeach stored at - 17#{176}Cntil u assayed. and the laboratory of the examinations

INTAKE 24-h urine

AND collection

HEALTH

STATUS

OF

VEGANS

2467

Medical At

nutritional investigation, the participants underwent a medical examination and completed a questionnaire regarding present and past nesses and diseases. All examinations were performed one ofthe authors (KB) and in Table 2 the main findings Statistics are summarized. The differences between Blood samples were taken for laboratory determinaStudents t test. tions. The analyses were performed using routine meth- with ods at the Department ofClinical Chemistry. University Hospital, Lund. Results and discussion High-density lipoprotein (HDL) cholesterol was determined by enzymatic cholesterol measurements ( 1 1 ) Nutrient in intake the supernatant obtained after precipitation of triglyceride-rich lipoproteins with dextran sulfate-MgC12 (12). The nutrient intake Low-density lipoprotein (LDL) cholesterol was calcu- given both quantitatively lated according to the formula described by Friedwald Table 3. As aqualitative et al. (13).

time

gracilis and Lactobacillus casei (ATCC 7488), respectively. as test organism. Total folate was obtained after enzymatic hydrolyses. Nitrate was reduced to nitrite with cadmium and determined spectrophotometrically and calculated from total nitrite minus the nitrite concentrasix tion obtained before the cadmium reduction. Oxalic acid was determined by a potassium permanganate titration illmethod. by

the

groups

were

calculated

the six vegans is and qualitatively in measure, the nutrient densities (intake/ 1000 kcal) have been A nahses offood and urine solids used. In these qualitatively terms the normal mixed diet has been shown to be rather simThe analyses of nutrients and other food components were performed as described previously (8, 9). In sum-ilar in different groups and independent of mary. energy intake expressed as kcal was calculated age and sex (9). The vegans were all middlefrom fat as 9.3 x g fat. from protein as 4.1 x 6.25 x g aged and a diet study of elderly was chosen nitrogen. and from carbohydrates as 4. 1 x g fat-free as reference mainly because the differences powder minus protein and 7% ash (carbohydrate by between these two groups regarding age were difference). The data were rounded off to the nearest 50 kcal. Fat was measured after extraction with chloroform probably of no significance. Recommended and evaporation. Fatty acid and sterol composition wereintakes of nutrients are for instance given in analyzed on lipid extracts using gas chromatography. similar amounts to all over5 1 yr. Protein was determined as nitrogen using a slightly Energy. The mean daily intake was 1700 modified Kjeldahl method. The amino acids were determined by GLC as trifluoracetyl-butyl-esters after acid kcal (7. 1 MJ) and 1400 kcal 5.9 MJ) for male ( hydrolysis (14). Tryptophan was determined with a specand female vegans, respectively (Table 3) trofluorimetric method (15) and methionine and cyswhen corrected for the dietary fiber included teine, by ion-exchange chromatography after performic the carbohydrate by difference. Correacid oxidation before acid hydrolysis (16). Low molecu- in figures for the 68- to 70-yr-old men lar weight carbohydrates were assayed both with enzy-sponding matic methods and with GLC. Glucose was determined and women on normal mixed diets were rewith a Tris-glucose oxidase reagent ( 17) and galactose spectively 2050 and 1600 kcal. with a galactose dehydrogenase-NAD system ( 18). In their reviews Sanders (7) and Miller and Fructose was assayed as described by Southgate (19). Mumford (22), reported that vegans have a Starch and sucrose were determined by measuring the lower energy intake when compared to subformation of glucose after enzymatic hydrolysis with amyloglucosidase and fI-fructosidase, respectively jects on common diets in agreement with the (Boehringer, Mannheim). Lactose was analyzed as gapresent study. On the other hand, previous lactose liberated after hydrolysis with /3-galactosidase. vegan studies have reported higher energy GLC of trimethylsilyl derivatives (20) yielded values in compared with the present study. Difvery good agreement with the enzymatic methods (N-G. intakes Asp, D. Birkhed, and C-G. Johansson, unpublished ferences in the food collection methods are data). The GLC method also gives sorbitol. Dietary fiberprobably the reason for such discrepancies. was determined gravimetrically after sequential digestion Considering the energy requirements of these with pepsin and pancreatin (21). Both water-soluble and groups men and women should consume water-insoluble components are included. The values age 2400 and 1800 kcal/day, respectively, accordgiven are corrected for the small residues of protein and starch remaining in the fiber fractions. Minerals and ing to Recommended Dietary Allowances trace elements were analyzed after wetashing in concen(RDA) (23). trated acids. Sodium and potassium were measured with The energy-yielding components expressed flame photometry. Iodine was analyzed by spectrophomin percentage of energy are shown in Table etry (the Sandell-Kotloff reaction). Vitamin B2 and folates were analyzed microbiologically using Euglena 3. In the vegan diet 29% originated from fat;

by

Downloaded from www.ajcn.org by guest on May 1, 2011

2468 TABLE Mean portions density

3

ABDULLA

ET

AL.

daily intake of energy and nutrients by three male and during 4 consecutive days: the nutrient quality of the is given and also compared with ordinary mixed diet
Daily Nutrients Volunteers )( diets) intake

three vegan

female diet

vegans expressed

each as

collecting nutrient

duplicate

Nutrient
.

densitvt Normal mixed 7) diet p1: values

Men

3 x

4)

Women

(3

x 4)

Vegan

diet

(6

X 4)

(35

Energy (kcal) Energy(MJ) Total fat (g)

1.700 250 7.1 11

55 12

1.400 5.9
56

370 1.6
18 31 6.7

(%)
Linoleic acid (g) Total sterols (mg) Cholesterol (mg) Protein (g) 31 424 29 49 (g)I 323 8 81 12 8

29 12 338 1 19 27 14 37 9 231
60

(%)
Total carbohydrate 49

(%) Sucrose (g)

Dietary fiber (g) Sodium(mmol) Potassium (mmol) Calcium (mg) Magnesium(mg)

43 12
62 9 11823 124 18 725 227 61590 17 6 13 2.3 3.9 0.8 6.9 3.6 82 29 0.39 0.08 620 120

Iron (mg)
Zinc (mg) Copper (mg) Selenium (jig) Iodine (JLg) Vitamin B12 (tg) Total folate (jsg) Other food components Oxalate (g) Nitrate (mg as NaNO:i) Nitrite (mg as NaNO:,)
a

7 43 9 7520 81 16 527 206 468 100 16 7 10 3.5 3.3 0.7 12.3 22.5 58 12 0.31 0.09 470 130

21

296 17 21 1 16 24 10 150 60 18 29 53 56 351 30029 9.0 6.5 2.0 5 39 0.19 301

4.6
52 7.2 2.4

1 14
8

7 405 4.3 2.3 184 90 140 67 30 6 12 2 105 12 44 47 5

7 4 12 8.4 135
2.4

21 6.3 49 30 392
6.5

10
2

<0.001 <0.001 <0.001 NS <0.001 <0.001 <0.001 <0.001 <0.001 NS

Downloaded from www.ajcn.org by guest on May 1, 2011

<0.00

17 6 94 1.9
1.5 12 57 55

11221
1.3 0.34 8 13 0.03 46 4.7 0.7 17 156 2.7 90

NS <0.001 NS <0.001 <0.001

<0.001

0.29

<0.001 <0.001 <0.001

<0.00 1

2.0

<0.001

0.98 0.22 169 51 2.4 3.2

0.84 0.22 158 38 3.3 2.9

0.5 0.18 92 22 1.7 1.3

0.16
18

0.1 8

1.7

1.1

<0.001 <0.001 NS

Mean

SD. per 1000 kcal, calculated diet versus normal for dietary fiber. including dietary on mixed fiber. the energy diet. values not corrected for dietary fiber.

t Amount :j:test t

vegan

Corrected By difference,

10% from protein; and 61% from carbohy- elderly, corresponding intake was 90 and 71 drates (by difference), with dietary fiber in- respectively. g, The fatty acid composition cluded. Corresponding values for the normal differed markedly from ordinary mixed diets. mixed diets were 40, 12.5, and 47.5%, respec- As shown in Figure 1, linoleic acid predomilively. nated in the vegan diet, comprising to 50 to fatty acids, whereas oleic and Previous studies on vegans (7, 22) have 60% of the total reported similar distributions of the energy- patmitic acids were the dominant fatty acids mixed diet. The ratio between yielding substances, except that these studies in the normal polyunsaturated and saturated fatty acids was showed a higher energy percentage from fat 3.7 in vegan diet, but only 0. 18 in the normal (35%) and a smaller one from carbohydrates mixed diet. According to the written food (55%). protocols, safflower oil was used in salads The mean food density (g/ml) was 0. 13 for the women and 0. 14 for the men, whichand warm dishes. should be compared with 0.20 obtained for Sterols. The daily intake oftotal sterols was on an average 424 and 338 mg for men and normal mixed diets (9). respectively, that was not signifiFat. Both the male and female vegans con- women, cantly different from the amounts found in sumed an average of 55 g fat/day. In the

NUTRIENT

INTAKE %)

AND

HEALTH

STATUS

OF

VEGANS

2469

The Swedish National Food Administration recommends a daily intake of 32 g protein/ 1000 kcal as a desirable level in planning 0 COMMON DIET diets, 28 g/1000 kcal as acceptable in evaluUVEGAN DIET 50 ation of diets and 24 g/l000 kcal as a minimum level (28). 1.0 According to RDA (23), a daily intake of 30 0.8 g protein/day and per kg body weight is 20 desirable. The mean consumption by the 10 women in the present study was 0.54 (DVI), 0.56 (DV-2), and 0.83 (DV-3) g/day per kg 12:0 1/.O 160 18:0 181 18:2 22:1 body weight. Corresponding figures for the male vegans were 0.83, 0.54 and 0.77, FIG. I. Fatty acid composition in the vegan diet three (U compared with the ordinary mixed diet (D) cxfor DV-4, Dy-S and DV-6, respectively. The pressed as percentages. Mean and SD. 67-yr-old subjects of our previous study consumed 0.69 g/kg (men) and 0.68 g/kg (women). the normal mixed diet when expressed per In terms of egg and milk protein, the Joint 1000 kcal. The sterol composition, however, FAO/WHO Expert Group consider 0.57 and was quite different (Fig. 2). In the vegan diet, 0.52 g/day/kg body weight, respectively (29), sitosterol was the main sterol component level for adults, which for the making up approximately 75% of the totalto be a safe man and women means a respective sterols, whereas cholesterol was less than 10%.reference daily intake of 37 and 25 g protein only. In the normal mixed diets, the corresponding The protein intake in the present study is values were 17 and 75%, respectively. Exinternational recommendations when pressed in nutrient densities, the vegan diets below origin is taken into account. Morecontained only about one-ninth as much cho- its plant over, the energy intake was low, which is lesterol as the normal mixed diet when exknown to influence protein requirements. Repressed per 1000 kcal. lating the protein intake to the urinary excreAlthough minor amounts of cholesterol tion of nitrogen (Table 4) the men, but not have been found in vegetable oils (24), it is the women, were in positive nitrogen balance possible that some of the material in the implying an adequate intake of protein as cholesterol fraction represents unidentified as energy. Vegans, deliberately depress sterols, reported in several vegetable foods well their protein intake because they believe that (25). Dietary plant sterols are poorly aba high dietary protein intake has deleterious sorbed in the intestine (26) and lower plasma cholesterol when administrated in large STEROL COMPOSITION (%) amounts. Thus the vegan diet contains little % saturated fat and cholesterol and is rich in polyenoic fat and plant sterols. Such a diet 0 COMMON DIET fulfills the criteria of a plasma-lipid-lowering a VEGAN DIET diet recommended by several committees (27). No rigorous scientific proof for the beneficial effects of a lipid-lowering diet is available at present. Protein. The mean daily intake of protein was 49 and 37 g for the male and female vegans, respectively (Table 3). Expressed in g/1000 kcal, the vegan diet averaged 24 g, which is significantly lower than for normal CHOLECAMPE STIGMA SITOmixed diet, amounting to 30 g/l000 kcal STEROL STEROL STEROL STEROL (Table 3). The lower protein intake by vegans FIG. 2. Sterol composition in the vegan diet compared to omnivores are in agreement with compared with the normal mixed diet (9) expressed previous studies (7, 22). percentages. Mean and SD.

FATTY

ACID

COMPOSITION

Downloaded from www.ajcn.org by guest on May 1, 2011

(U) as

2470 TABLE Comparison and three 4 of female intake and urinary vegans collected food excretion and

ABDULLA

ET

AL.

urine

of nitrogen, during

sodium, and 4 consecutive

potassium: days
Urinary

three

male

Dietary Men Women

intake All Men

excretion Women All

Nitrogen Sodium day) Potassium day)

.

(g!day) (mmol! (mmol! (SD). versus

7.8 118 124

(1.3) (244 (19)

5.9 (l.6)t 75 (21) 81 (17)

6.8 96 103

(1.7) (31) (28)

6.8 82 86

(1.3) (25) (28)

7.6 88 86

(1.5) (35) (26)

7.2 85 86

(1.5) (30) (26)

Mean
t

test, intake :1: < 0.001. p

excretion

0.001< p < 0.01.

p<O.OS.

TABLE 5 The daily intake of essential comparison corresponding intake of the essential amino

amino acids, calculated data on ordinary mixed acids for adults

Vegan Men diet Ordinary Male

as diets

mgikg body wt per and the recommended

day

and

for

Downloaded from www.ajcn.org by guest on May 1, 2011

mixed pensioners

diet

Vegan Women

diet

Ordinary Female

mixed pensioners

diet

Recommendaiionsr

Isoleucine Leucine Lysine Total serum-containing Total aromatic Threonine

Tryptophan
Valine
*

23 29 27 37 35 21 11 26 range.

(17-31) (23-44) (21-28) (22-40) (29-42) (18-24) (9-17) (21-32)

28 55 34 30
42

(19-52) (36-95) (19-57) (23-41)

(34-60)

15 26 21 32
24

(14-24) (24-41) (18-31) (32-38)

(21-44)

25 48 30 28
40

(13-38) (23-74) (14-76) (24-33)

(30-47)

12 16 12 10
16

23(11-39) 8 (3-15) 32 (21-56)

14(14-23) 10 (9-13.5) 19 (18-29)

25(12-38) 8 (3-14) 29 (17-41)

8 3 14

Median RDA

and (23).

do not use refined sugar, their intake of sueffects on the health. The health effects of the crose, expressed as g/ 1000 kcal was only 4 g low protein intake are further discussed in lower when compared to normal mixed diets, relationship to clinical and biochemical rewas a much smaller difference than sults (see Clinical and biochemical findings which was expected. There was, however, a striking below). The intake of essential amino acids by the difference between male and female vegans vegans was calculated as mg/day and with kg regard to daily intake of sucrose, male vegans consuming twice as much as female body weight and compared with correspond43 and 2 1 g sucrose per day, respecing values for the elderly subjects on a normal vegans, mixed diet (Table 5). The intake of essential tively. According to food balance sheets, the suamino acids was found to exceed the recomconsumption per capita in Sweden is mendations set by the Food and Nutrition crose Board [RDA (23)]. It seems that food proteins 120 g/day, which corresponds to about 40 g/ In our studies on mixed diets, less of plant origin could supply all the essential 1000 kcal. than half of that amount was recorded (8, 9). amino acids when proteins from pulses are reason for this discrepancy is unclear but combined with protein from cereals and other The a few explanations have been tendered. The vegetables (7). may have consumed less sugar Carbohydrate. The mean daily intake of volunteers during food sampling days than usual, the total carbohydrate (by difference) was 323 participating in the study may be low and 23 1 g for the male and female vegans, subjects sugar consumers and/or some of the sugar respectively. The carbohydrate composition be hydrolysed (inverted) during prepain vegan diets compared with normal diets may is of meals. The comparatively high shown in Figure 3. Even though the vegans rations

NUTRIENT g/1000 kcol 0

INTAKE

AND

HEALTH

STATUS

OF

VEGANS

2471

COMMON

DIET
DIET

50

#{149} VEGAN

tI1j

40 30

:
GLUCOSE
FIG. in g!I000 3. Carbohydrate kcal (nutrient

(Ii
FRUCTOSE
in

SUCROSE
the vegan and SD.

LACTOSE
diet (U)

STARCH

DIETARY

SORBITOL

FIBER composition density). Mean compared with the ordinary mixed diet (El) expressed

Downloaded from www.ajcn.org by guest on May 1, 2011

calculations were based on crude fiber values. amounts of sucrose in the vegan diet may be During recent years more reliable techniques traced to fruits and berries consumed for have been developed for analyzing dietary breakfast as well as to vegetables, such as fiber (2 1), which also include the water-solcarrots, peas, etc. uble part of dietary fiber. The intake of monosaccharides, as with and trace elements. Little inforglucose and fructose, was 3 times higher than Minerals is currently available regarding both in mixed diets. This may also be due to fruits mation and toxic minerals and trace eleand might partially be a result of enzymatic essential in vegan diet, except for iron (7). On or acid hydrolysis of sucrose or starch during ments the other hand, the levels in the diet per se do preparation and storage of vegetables and not provide information concerning the availfruits. The starch content was somewhat lower ability of these nutrients. Here, the medical than in the mixed diets, probably due to the and biochemical investigation ofeach subject fact that the vegans did not use bread fromin the present study may give valuable information (see below). cereals. Sorbitol constituted 5.4 2 g/l000 kcal. The mean daily intake of minerals and Nine or 10 of the mixed diets earlier studied, trace elements are shown in Table 3. Excontained only traces ofsorbitol (<0.7 g/l000 pressed in nutrient densities and compared kcal; D. Birkhed et al., unpublished obser- with normal mixed diets, the vegan diet convations). The rather high sorbitol concentratamed similar amounts of calcium and sotion corresponding to a daily intake of about dium 346 versus 394 mg/l000 kcal and 53 10 g/day is still below the level at which versus 49 mmol/l000 kcal, respectively. This was surprising because the vegans did not gastrointestinal problems occur. As expected, the vegan diet contained prac- consume milk or dairy products and did not tically no lactose. The vegan diet is a high use any table salt. bulk diet. The concentration of dietary fiber/ The potassium content of the vegan diet 1000 kcal was about 5 times higher in the was nearly doubled (56 versus 30 mmol/l000 vegan diet (29 g/l000 kcal) as compared with kcal), and the magnesium was approximately normal mixed diets (6 g/l000 kcal). The times 3 higher (300 versus 1 12 mg/bOO kcal) mean daily intake of dietary fiber was 75 andwhen compared with normal mixed diets. In 52 g for men and women, respectively. Hard- addition the vegan diet contained about 1.5 inge and Stare (1) estimated that vegan diets times higher concentrations of iron (9.0 vercontained on an average 2#{189} times as much sus 6.5 mg/bOO kcal), and zinc (6.5 versus dietary fiber as normal mixed diets. Their 4.7 mg/l000 kcal), but 3 times higher concen-

2472

ABDULLA

ET

AL.

trations of copper (2.0 versus 0.7 mg/l000 cate portions contained respectively 33, 47, kcal). The higher amounts ofcertain minerals and 66 jig. and trace elements may be because the vegan It is known that food items ofanimal origin diet comprises of fewer refined food items, as for instance, fish, shellfish, and viscera are and that many vegetables and whole grain the richest sources of selenium, whereas Cecereals are rich in minerals and trace elereals, vegetables, etc., are poor sources (32). ments. Further, it has been demonstrated that SwedA comparison of the daily intake of these ish soils, like those in all the Scandinavian nutrients with recommendations given by countries, are low in selenium (32). No recRDA (23) showed that the recommendation ommendation is presently available regarding for magnesium (300 mg) was exceeded. Fur- the dietary needs of selenium in man, but the thermore the male vegans consumed moredaily requirements have been estimated to be than 10 mg iron daily, while the female ye100 to 150 jig (33). Normal mixed Swedish gans came very close to the recommended diets will supply a median intake around 25 daily intake of 18 mg. Although the vegan jig/day (9), with a range from 10 to 95 jig. In diet was richer in zinc, the daily intake did recent a review of several studies on daily not reach the recommended 15 mg for either dietary intake of selenium in different counmale or female vegans. According to RDA tries ranges from 56 jig/day (New Zealand) (23), adults should consume 800 mg calcium to 326 jig/day (Venezuela) were estimated daily. On the other hand, FAO/WHO (30) (34). recommends a daily intake of 400 to 500 mg Vitamins. In the present study folate and which was exceeded by the vegans. vitamin B12 were the only vitamins measured. The folate content averaged 3 times higher in The vegan diet was low in iodine, only 39 vegan diets than in the normal mixed .tg/l000 kcal versus 160 tg/kcal in normal the mixed diets. Iodine occurs mainly in fortified diets, (301 versus 90 jig/l000 kcal), which expected, since especially green vegetatable salt, marine fish, milk, and milk prod- was are important sources of folate. RDA ucts, which were not consumed by the vegans. bles (23) recommends a daily intake of jig 400 The iodine requirements have been estimated was exceeded by the vegans in the to be 1 tg/kg body weight per day (3 1). Five which study. According to FAO/WHO (35) of the six vegans reached this level, but DV-present intake of at least 200 is sufficient. jig 1 did not. According to RDA (23), a dailya daily daily intake of vitamin B12 was about intake of 90 Lg iodine is recommended for The is approximately onewomen and 125 tg for men in these ages.0.3 to 0.4 jig, which of that in normal mixed diet. Vitamin These levels were not reached by the vegans tenth B12 is a product of microbiological synthesis in the present study, but no clinical signs of and is not found in plant foods and should iodine deficiency were observed. Considering therefore not be present in vegan diets. Analthese low iodine intakes, one must keep in yses of different vegan food items, including mind that the diet was collected at F#{246}llingevegetables, yielded vitamin B12 garden, which is situated in an area where fermented between 10 and 70 ng/ 100 g wet the occurrence of endemic goiter was highvalues weight, which very well may explain the daily before iodine fortification of table salt was intake by the vegans. The small amount of introduced. B12 recovered in the vegan diet may Also selenium was lower in the vegan diets vitamin been due to contamination with B12averaging only 5 .tg/l000 kcal compared with have microorganisms present in food. 18 .tg/lOO0 kcal in normal mixed diets. Onproducing average, the female vegans consumed 6.9 tg RDA (23) and FAO/WHO (35) recoma daily intake of 3 and jig 2 of vitamin selenium/day. Among the duplicate daily mends B12, respectively, but according to Herbert portions collected by the women, one con(36) as little as 0. 1 jig/day meets the minimum tamed undetectable levels, i.e., below 3 jzg/ requirements. Vitamin 12 deficiency is the daily portion. The highest concentration of common nutrient deficiency described selenium found was 13.3 zg/daily portion. most Among the 12 daily duplicate portions colamong vegans (37-41). Recently, the serious lected by the male vegans, nine contained no consequences for babies of vegan mothers been described (42). detectable selenium. The other three dupli- have

Downloaded from www.ajcn.org by guest on May 1, 2011

NUTRIENT

INTAKE

AND

HEALTH

STATUS

OF

VEGANS

2473

intake of dietary fiber might be the explanaNitrate-nitrite. The vegan diets contained for the absence of overweight but the 92 mg nitrate/l000 kcal versus 18 mg/bOO tion exercise included in the vegan kcal in normal mixed diets. On the otherrole of physical life style should not be dismissed. hand, the nitrite levels were equal in both blood pressures of the vegans in the types of diets, being 1.5 to 2 mg/l000 kcal. The present study were low considering their ages The nitrate intake expressed per kg body (Tables 2). A high salt intake and overweight weight per day varied between 1.6 and 3.4 has been reported to be associated with high mg and the nitrite intake between 0.03 and 0. 10 mg/kg body weight per day. The acblood pressure. Obviously, these two risk facby the vegan lifestyle. ceptable daily intake ofnitrate as NaNO3 andtors are reduced consequence of the low fat diet nitrite as NaNO2 is 5 and 0.2 mg/kg body Another high in bulk and polyunsaturated fatty acids weight per day, respectively (43). These acbe the low concentrations of plasma ceptable daily intake bevels were not exceeded may lipids also reported by Sanders (7). As seen by the vegans. 6, the concentrations of cholesterol, Oxalate. The vegan diet contained 3 times in Table HDL, and LDL concentrations higher concentrations of oxalate compared triglyceride, with normal mixed diets. Almost 1 g/day was were in the lower half of the reference range. consumed by the vegans. The mean ratios LDL-cholesterol/HDL-cholesterol were 1.8 (men) and 3.1 (women). The Validity of duplicate portion technique former values agree with findings in 30- to 40-yr-old vegetarians (44), whereas our ratio To check the validity ofthe food collections the women was considerably higher. It using the duplicate portion sampling tech-for should be emphasized that the reference nique, the urinary excretion of sodium, poranges for the lipoproteinclasses in Table 6 tassium, and nitrogen was monitored and for age; these 49- to 55-yrcompared with corresponding intake values. are not corrected The male vegans, but not the females, exold subjects could be expected to have LDL values in the upper reference range. A low creted significantly smaller amounts of soLDL/HDL cholesterol ratio has been sugdium and potassium (p 0.001) < when comto be advantageous from the atheropared to their intakes. Regarding nitrogen, gested point ofview (45) and may also reflect there was no significant difference between genetic preferential use of endogenous rather intake and excretion by the male vegans. Thethe women, however, excreted more nitrogen than exogenous fat. Plasma prealbumin was low in DV- 1 and (0.01 < p < 0.001) than they consumed. The losses of nitrogen through the skin and in the DV-2 and borderline in DV-3, but was norfor the male vegans (Table 6). All six faeces are not included in the excretion values mal tabulated in Table 4. These results indicate subjects had plasma albumin values near the that the women but not the men underesti- borderline value of 38 g/l. This was probably mated their food collections. Correlation due to a low protein intake or a low protein/ studies between intake and urinary excretion energy ratio in the vegan diet. On the other showed no significances for any of these nuhand, none of the subjects studied presented trients. clinical or biochemical indications of protein deficiency. Their retinol-binding protein and Clinical and biochemical findings ceruloplasmin for example were all normal Evaluation of the vegan diet may be made (Table 6). by comparisons with accepted nutrient re- Two subjects had blood folate levels above quirements and recommendations. Another normal (Table 7). Serum cobalamins were approach is to relate it to the clinical and analyzed both with the radioimmuno assay biochemical status of vegans. technique and microbiologically using . graE It has been reported that vegans have lower cilis as test organism. One male vegan (DVbody weights than persons on normal mixed 4), having been a vegan for the longest period diets (3). None of the vegans in the present of time, 10 yr, had a borderline value using study was overweight, which is noteworthy in the radioimmuno assay technique but was this age group (Table 2). The low intake of deficient in the microbiological test, while the energy and fat in combination with a highother five subjects had serum cobalamins

Downloaded from www.ajcn.org by guest on May 1, 2011

2474

ABDULLA

ET

AL.

a > 0 sO r 00 0
s

0 if5

.10 sO 00000

O5000IsO

sO evi

if5if

sOeF5O 0 1 0 0
-

(510
rl 0
-

00 0

sO

es4

Qs

r-

00 00
-rn-

* 0 0

1-O

00rO5 0 0
(

r.4 0 ( Cv

50 ifflif 0 0
-

0 (54 (54 0
(5-

00
-

ris. 0

oc0
(54 0
-

0 0 sO

0 a > 00 00 r .1sO
(

(54 000ifrsI0-

(5-

0
-

0 sO (54 rnor-40 (54

if5 00if-sO

r-

-(54r4

(5-

> 0

rsl

ri -

.. (i

00

* 0

1-ifs if

Q
(54

en
_

oo 00

.sOifs
Cl

(54

rfl

ov

0
00

rs0 0 0 r 0 sO 0

0 0 00(54.0

sO 00

Downloaded from www.ajcn.org by guest on May 1, 2011

Iri 00
-

(5-

00 sO
,,.

0 0 (500 (5- 0 r
-

>

ri

00

sO

* 0

1e if

sO

ri

Ov

00 r if if if

0-r-4r

(i

>

00 (54

(54

r
fs

* 0
00

100
-

00 (54

if sO

05

sOO

r
sr-0-

0
-

-i-.

I-.

r -

rnO. 0

s0 (4 0
if

0
(54

I-

0
(54

0 (-4 > 0(4 (t 00

if

Nfs 0 00 (51(5( 0 : fs #{176} i 0

00 (540
-

1
0 0
-

sO

(5lsO

-r-

0 ( (-4

555

:
E
#{176} m

:
on . .
55U n
5-

0.

on
a a
0

:E
on
=.

-aUU

_5
-a
#{176}

C
a.

a -5.-c

.,

Ifl
.-

d --a i!I
Ecu-a
U

.n.E
U sO 0

a--oa a

.
>

.-

a..... .c
0

HD EE

#{176}H
.-ll
-.-

i-nni

-E
ca,,5

EE
,,

cucucu
ini#,i#i<

EEu0
nnc..<

O.r/)C/
.

nn

e
..

UC/C

NUTRIENT

INTAKE

AND

HEALTH

STATUS

OF

VEGANS

2475

r > (40ifs0
-

0 0-sO

if (5-0
-

r r1

(54 > (54

-

0
-

0. 0 O (59
00

Ov

00

-V -

(5

sO r-4rI0*fs if5000r0
(5..

(54 (-4 0-

00

(54

0ri

I
0-.--.ls0(-.000 00 00 (-4 s0 m

>!

(54/sO0 r-

r1 if > Os 0(59 0--sO00 ( 0 0 sO

1--

(54000 Os

(-4

00 0 -0 Os if 0 (5-4 00 ri 4 N
-

::

r1

sCV

00
ifs

00 Os 0

.-.irs

a on a
0
0

U 0

C
a.

a
0

on a
a
U i.e

within the normal range. Usually, man has vitamin B12 stores to meet the daily requirements for about 3 yr and when small amounts of the vitamin is ingested daily it may take 5 to 10 yr or even longer before clinical signs of deficiency appears (7, 40). The Hb-values were all within the normal range indicating sufficient absorption of the dietary iron. One had subnormal serum iron with normal transferrin and serum-ferritin values. Five of the six vegans had subnormal serum total iron-binding capacity. Despite the low iodine intake, no clinical or laboratory parameter indicated any sign of iodine deficiency. S-TSH and T3 values did not indicate hypothyroidism. The low intake of selenium among the vegans found in the present study is hard to evaluate. Specific symptoms of selenium deficiency in man is rarely reported. Recent studies from China indicate that dietary selenium deficiency may be associated with cardiomyopathy in chibdren (46). Supplementation with 0.5 mg sodium-selenite/week in children 1 to 5 yr old and 1 mg/wk in children 6 to 9 yr old reduced the prevalence of the disease markedly. Selenium is an essential constituent of erythrocyte glutathione peroxidase in several animal species (33). It may play a role also in the plasma transport of tocopherol. At present, attention is being directed to the association of selenium deficiency with malignant diseases (47). Bilirubin was increased in subject DV-4 and serum aspartate aminotransferase was increased in subject DV-2 without any other known hematological or hepatic disease. Creatinine clearance was subnormal in five of six vegans. Serum creatinine values were normal, however. The urinary excretion of calcium and sodium (Tables 4 and 6) was lower than the normal range, whereas potassium was excreted in amounts above the normal range. This could be expected from their dietary habits.

Downloaded from www.ajcn.org by guest on May 1, 2011

an
Un

>

Conclusion
.

:
U

on
>* .t-

.EEE8 . U U U Oononon00m

.
U

The fered diet. linoleic cium,

nutrient quality of the vegan diet difin several respects from normal mixed For some essential nutrients such as acid, the essential amino acids, caliron, magnesium, potassium, copper,

2476

ABDULLA

ET

AL. mination of total cholesterol in serum. J Clin Chem Clin Biochem 1974:12:226. Danielsson B, Ekman R, Fex G, et al. Changes in plasma high density lipoproteins in chronic male alcoholics during and after abuse. Scand J Clin Lab Invest 1978:38:113-19. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of loW-density lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge. Clin Chem 1972; 18:499-502. Nair BM. Gas-liquid chromatographic analysis of amino acids in food samples. J Agric Food Chem
1977;25:6 14-20.

and folic acid the recommendation set by either RDA (23) or FAO/WHO (29-31) were 12. met, while for other nutrients like energy, total protein, zinc, iodine, and vitamin B12 these recommendations were not met. In ad13. dition the selenium intake was very low, but the clinical significance is not clear at present. Clinical and biochemical data of the yegans revealed no symptoms of nutritional 14. deficiencies, except for one subject with a subnormal serum vitamin B12. 15. Also other dietary components differed in vegan diet compared to normal mixed diet, for instance the vegan diet contained a high16. polyunsaturated/saturated fat ratio, a low 17. concentration of saturated fat and cholesterol as well as high concentrations of dietary fiber and plant sterols, compared to normal mixed 18. diet which probably contributed to certain clinical and biochemical findings in the ye19. gans, e.g., low levels of blood lipids, low blood pressures, and normal to low body weights for their age. 20. References
Hardinge etarians. 2. West ED, veganism, controls. 39 1-7. 3. Ellis FR, ings and
249-55.

Oste R, Nair BM, Dahlqvist A. A simple method for determination oftryptophan in food samples. J Agric Food Chem 1976:24:1141-4. Moore S. On the determination ofcysteine as cysteic acid. J Biol Chem 1962:238:235-7. Dahlqvist A. Determination of maltose and isomaltase activities with a glucose oxidase reagent. Biochem J l961;80:547-51. Galactose and lactose. Laboratory lnformation. Mannheim, West Germany: Boehringer CF & Soehne, 1967. Southgate DAT. Determination of food carbohydrates. New York: Applied Science Publishers Ltd. 1967:113-14. Fuchs G, Gawell B-M, Lidhem B-M. Quantitative determination of low-molecular carbohydrates in foods by gas-liquid chromatography. Swed J Agric Res l974;4:49-52. Asp N-G, Johansson C-G. Techniques for measuring dietary fibre. Principal aims of methods and a comparison of results obtained by different techniques. In James WPT, Theander 0, eds. The analysis of dietary fibre in food. New York: Marcel Dekker Inc, 1980:173-89. Miller DS, Mumford P. The nutritive value of Westem vegan and vegetarian diets. Plant Foods Human Nutr 1972:2:201-13. Recommended Dietary Allowances, 8th ed. Food and Nutritional Board, National Academy of Sciences. Washington DC: National Research Council, 1974. Johansson A. The content and composition of sterols and sterolesters in sunflower and poppy seed oils. Lipids 1979; 14:285-91. Weihrauch JL, Gardner JM. Sterol content of foods of plant origin. J Am Diet Assoc 1978:73:39-47. Kuksis A, Marai L, Myher JJ. Geher K. Identification of plant sterols in plasma and red blood cells of man and experimental animals. Lipids 1976:11:58 16. Dietary fats and oils in human nutrition. Rome: FAO, 1977:32. Kost och Motion. Socialstyrelsens n#{228}mnd for halsoupplysning, Stockholm. 1975:21-6 (in Swedish). FAO/WHO: Energy and protein requirements. Report of a joint FAO/WHO ad hoc Expert Committee, WHO Technical Report Serie no. 522. Geneva: WHO, 1973. FAO/WHO. Calcium requirements. Report of a FAO/WHO Expert Committee on calcium requirements. WHO Technical Report Series no. 230. Geneva: WHO, 1962.

Downloaded from www.ajcn.org by guest on May 1, 2011

1.

MG, Stare FJ. Nutritional studies Am J Clin Nutr 1954:2:73-88. Ellis FR. The electroncephalogram vegetarianism, vitamin B12 deficiency J Neurol Neurosurg Psychiatry Montegriffo investigations. VME. Am Veganism, J Clin

of in

21. veg-

and 1966:29:

clinical find- 22. Nutr 1970:23:

4.

5.

6.
7.

8.

9.

23. Sanders TAB, Ellis FR, Dickerson JWT. Studies of vegans: the fatty acid composition of plasma choline phosphoglycerides, erythrocytes, adipose tissue and breast milk and some indicators of susceptibility to ischemic heart disease in vegans and omnivore con-24. trols. Am J Clin Nutr 1978:31:805-13. Sirtori CR, Agradi E, Conti F, Manter 0, Gatti E. 25. Soybean-protein diet in the treatment of type-Il hyperlipoproteinemia. Lancet 1977; 1:275-7. 26. Ellis FR, Sanders TAB. Angina and vegan diet. Am Heart J l977;93:803-5. Sanders TAB. The health and nutritional status of vegans. Plant Foods Man 1978;2:18l-93. Borgstrom B, Nord#{233}n Akesson A, B, Jagerstad M. A 27. study of food. Consumption by the duplicate portion 28. technique in a sample of the Dalby population. Scand J Soc Med 1975;(suppl lO):l. Borgstr#{246}m B, Nord#{233}n , Akesson A B, Abdulla M, 29. J#{228}gerstad M. Nutrition and Old Age. Chemical analyses of what old people eat and their states of health during 6 years of follow-up. Scand J Gastroenterol30. ogy l979:l4(suppl 52):l. Johansson

10.

Tidskriften
1 1 . Roeschlau

L. Vital mat. 4 ed. En receptbok fOr H#{228}lsa. iding#{246}, 1976 (in Swedish). L P. Bernt E, Gruber W. Enzymatic

fran deter-

NUTRIENT 31. Food and Nutritional Board:

INTAKE Iodine nutrition

AND

HEALTH in the

STATUS

OF

VEGANS

2477

32.

33. 34.

Washington DC: National Academy of Sciences, 1976. 40. Underwood EJ. Selenium. In: Trace elements in human and animal nutrition, 4th ed. New York: Academic Press, 1977:302-6. 41. Burk RF. Selenium in nutrition. World Rev Nutr 42. Diet 1978,30:88-106. Levander OA. In: Selenium-Tellurium in the environment. Pittsburgh. PA: Industrial Health Foun43. dation, 1976:26-36. FAO/WHO Expert Group. Requirements of ascorbic acid, vitamin D, vitamin B12, folate and iron. WHO Technical Report Series no. 452. Geneva: WHO, 1970. 44. Herbert V. Nutritional requirements for vitamin B2 and folic acid. Am J Clin Nutr 1968:21:743-52. 45. Sanders TA, Ellis FR, Dickerson JWT. Haematological studies on vegans. Br J Nutr 1978:40:9-15. Higginbottom MC, Sweetman L, Nyhan WL. A syndrome of methylmalonic aciduria, homocysteinuria, megaloblastic anaemia and neurological abnor- 46. malities in a vitamin B2 deficient breastfed infant of 47. a strict vegetarian. N EngI J Med 1978:299:317-23. Zmora E, Gorodischer R, Bar-ziv J. Multiple nutri-

United

States.

35.

36.
37. 38.

39.

tional deficiencies in infants from a strict vegetarian community. Am J Dis Child 1979:133:141-4. Frader J, Reibman B, Turkewitz D. Vitamin B12 deficiency in strict vegetarians. N Engl J Med 1978:299:1319. Anonymous. Vegetarian diet and vitamin B12 deficiency. Nutr Rev 1978;36:243-4. Anonymous. Vitamin B,2 deficiency in the breastfed infant of a strict vegetarian. Nutr Rev 1979:37:1424. FAO/WHO. Toxicological evaluation of certain food additives with a review of general principles and of specifikations. FAO Food Nutrition Series no. 53. Geneva: WHO, 1974:17. Burslem J, Schonfeld G, Howald MA, Weidmans SW, Miller JP. Plasma apoprotein and lipoprotein levels in vegetarians. Metabolism 1978:27:7 1 1-719. Gordon T, Castelli WP, Hjortland MC, et al. High density lipoprotein as a protective factor against coronary heart disease: the Framingham study. Am J Med 1977:62:707-14. Selenium in the heart of China. Lancet 1979,2:88990. Schrauzer GN. Selenium and cancer: A review. Bioinorg Chem 1976:5:275-8 1.

Downloaded from www.ajcn.org by guest on May 1, 2011