Content uploaded by Naomi K Fukagawa
Author content
All content in this area was uploaded by Naomi K Fukagawa
Content may be subject to copyright.
524 Am J C/mn Nuir 1990:52:524-8. Printed in USA. © 1990 American Society for Clinical Nutrition
H igh-carbohydrate ,high-fi ber diets increase pen pheral insulin
sensitivity in healthy young and old adults13
Naomi K Fukagawa, James WAnderson, Geja Hageman, Vernon R Young, and Kenneth L Minaker
ABSTRACI’ To examine extraalimcntary effects of high-
carbohydrate, high-fiber(HCF) diets, insulin-mediated glucose
disposal employing the euglycemic clamp and hepatic glucose
output (HOO) employing [6,6-2H2]glucose were measured in
12 healthy young and old individuals before and after 21-28 d
ofan HCF diet. Diet lowered fasting concentrations of glucose
from 5.3 ± 0.2 to 5.1 ± 0.1 mmol/L (p <0.01) and insulin
from 66.0 ± 7.9 to 49.5 ± 5.7 pmol/L(p <0.0 1). Fasting serum
cholesterol decreased from 5. 17 ± 0. 18 to 3.80 ± 0.20 mmob/L
(p <0.01) in young individuals and from 6.15 ± 0.52 to 4.99
±0.49 mmol/L (p <0.01) in elderly individuals. Fasting serum
triglyceride concentrations, basal HOO, and insulin suppres-
sion of HGO were unchanged by the diet. Glucose disposal
rates increased from 18.87 ± 1.66 before to 23.87 ± 2.78
mol. kg’ .min after the diet (p <0.02). Therefore, HCF
diets may improve carbohydrate economy by enhanced pe-
ripheral sensitivity to insulin. Am J C/in Nuir l990;52:
524-8.
KEY WORDS HCF diet, insulin sensitivity, aging
Introduction
In healthy individuals high-carbohydrate, high-fiber (HCF)
diets lower fasting plasma glucose and insulin concentrations
as compared with the usual free-choice diets (1, 2). In clinical
practice, HCF diets have been advocated as an adjunct to the
management of diabetes mellitus because they lower insulin
requirements and improve glycemic control in lean diabetic
individuals without altering body weight (3-8). In addition to
their effects on carbohydrate metabolism, HCF diets lower
plasma cholesterol (1 ,6, 7) and may reduce the risk of bowel
cancer (9). Thus, recommendations were made for the general
population to increase their intake of complex carbohydrate
and dietary fiber (10). Aging in all mammalian species exam-
med was shown to be associated with a progressive deteriora-
tion in glucose tolerance (1 1,12). Studies showed that this is
due, in part, to diminished peripheral-tissue sensitivity to insu-
bin (13-16).
The mechanisms responsible for the metabolic effects of
HCF diets have not been clarified (7). Effects of intestinal, he-
patic, and peripheral tissues may contribute to the effect of
HCF diets on glucose metabolism. To evaluate the hypothesis
that HCF diets may induce metabolic changes beyond those
occurring in the gut and liver and to determine whether HCF
diets may alter age-related changes in insulin sensitivity, we
evaluated peripheral-tissue insulin responsiveness before and
after an HCF diet.
Subjects and methods
Subjects
Six healthy young male students (age range 18-24 y, I
± SEM 21 ± 1 y), with 96-I 32% relative body weight (17), and
six healthy elderly men and women (age range 67-86 y, I
±SEM 72 ± 3 y), with (89-132% relative body weight, partici-
pated in the studies. All had normal hepatic and renal function
tests. None had a family history ofdiabetes melbitus and none
was taking medications on an acute or chronic basis. All had
maintained stable body weights in the previous 6 mo. Oral glu-
cose (40 g/m2) tolerance tests were normal. Specifically, elderly
subjects did not demonstrate impaired glucose tolerance as de-
fined by the National Oroup Diabetes Study (18). All subjects
were carefully informed of the nature, purpose, and possible
risks of the studies before they gave written consent to partici-
pate. The protocol was approved by the Committee on the Use
of Humans as Experimental Subjects at the Massachusetts In-
stitute ofTechnology (MIT).
Experimental design
The subjects were studied, in sequence, while consuming
their usual ad libitum diet (period 1) and after consuming a
weight-maintaining HCF diet for 21-28 d (period 2). Four of
the six young men returned 2-6 wk after resuming an ad libi-
IFrom the Clinical Research Center and Department ofApplied Bi-
ological Sciences, Massachusetts Institute of Technology, Cambridge:
the Charles A Dana Research Institute and the Harvard-Thorndike
Laboratory, Department of Medicine, Beth Israel Hospital and Har-
yard Medical School, Boston; the Geriatric Research Education and
Clinical Center, Brockton/West Roxbury VA Medical Center, Boston;
and the VA Medical Center and Department of Medicine, University
ofKentucky, Lexington, KY.
2Supported by the HCF Nutrition Research Foundation and a grant
from the American Federation for Aging Research. NKF was sup-
ported by NIH training grant AM 7070 and a Clinical Associate Physi-
cian Award, NIH GCRC grant 2M01-RR00088. KLM was supported
by a Greenwall Foundation Award from the American Federation for
Aging Research and NIH grant AGO 0599.
3Address reprint requests to NK Fukagawa, Beth Israel Hospital,
330 Brookline Avenue, Boston, MA 02215.
Received August 21, 1989.
Accepted for publication November 1, 1989.
by guest on July 10, 2011www.ajcn.orgDownloaded from
HCF DIET AND INSULIN SENSITIVITY 525
TABLE 1
Estimates ofdaily nutrient intakes in young and old subjects*
Control diet HCF diet
Young Old
Young Old
Protein,total(g/d) 136 ±17 86 ± 6 138 ± 6 80 ± 4
Carbohydrate,total(g/d) 304 ±30 218 ±lb 538 ±22 305 ±30
Simple 174 ±31 109 ± 8 278 ± lb 153 ± 18
Complex 134 ±12 108 ± 9 260 ± 12 152 ± 15
Fat,total(g/d) 143 ±19 80 ± 10 49 ± 2 29 ± 2
Saturatedfattyacids 58 ±932 ±5 12 ±1 7 ±I
Monosaturated fatty acids 54 ±6 23 ±3 19 ± I 8 ± 1
Polyunsaturated fatty acids 26 ± 7 17 ±1 18 ±1 7 ±1
Cholesterol (mg/d) 755 ±9 1 720 ± 74 134 ± 6 90 ±3
Plantfiber(g/d) 15.5± 2.7 18.5± 1.9 88.2± 11.1 67.7± 5.9
Insoluble 12.0 ±2.4 14.8 ± 1.7 64.6 ±8.0 50.6 ±5.1
Soluble 3.5± 0.6 3.7± 0.3 23.6± 3.1 17.1± 3.1
Totalenergy(kcab/d) 3059 ±152 1917 ± 152 3138 ±125 1741 ± 154
tum, self-selected diet for a third study. At the end of each di-
etary period, a 3-h 10 mU.m2.min’ euglycemic insulin-
clamp study (1 3-16) was conducted to permit quantification
of the effects of steady-state physiologic hyperinsulinemia on
glucose disposal and hepatic glucose output. For 150 mm be-
fore and throughout the first two clamps, a primed constant
infusion of [6,6-2H2]glucose was used to estimate hepatic glu-
cose production. Blood glucose, cholesterol, triglyceride, and
insulin concentrations were measured weekly during the HCF
diet period by use ofpreviously described methods (6).
Diet
In the young subjects the usual ad libitum (control)diet com-
position was determined from a 2-wk dietary recall. The HCF
diet was prepared in the metabolic kitchen at MIT and con-
sumed by the subjects under supervision. The elderly subjects
were admitted to the MIT Clinical Research Center for the du-
ration ofthe study. They consumed a weight-maintaining diet
for 3 d before the first insulin-clamp study and remained in the
CRC during the 4-wk HCF diet period.
The HCF diet (Table 1) provided -‘--68% ofenergy from car-
bohydrate as compared with ‘-43% for the free-choice or con-
trol diet. Fat intake was estimated to be less during the HCF
diet period than during the ad libitum diet periods (42% vs 14%
of total energy); protein intakes were similar (18%). Both
groups of subjects received diets that were very high in fiber
compared with the average intake of American adults, esti-
mated at ‘-6 g/l000 kcal or 10-18 g/d(19). Total dietary fiber
intake in the HCF diet was 33 g/l000 kcal as compared with
an estimated 7 g/l000 kcab in the control diet. For the young
subjects diets were designed to provide 35 g/l000 kcal but
not to exceed 100 g/d. For the elderly subjects diets were de-
signed to provide 70 g/d or 35 g/l000 kcal. No sucrose was
added to the HCF diet; simple carbohydrate refers to monosac-
charides, disaccharides, and trisaccharides contained in milk,
grain products, vegetables, and fruits. The plant fiber was pro-
vided by whole grain or grain cereals and breads (40%); vegeta-
bles such as beans, corn, or peas (20%; other vegetables (31%:
and fruits (9%). The nutrient, energy, and fiber content were
calculated from food composition tables as previously de-
scribed (20). Water-soluble fibers included pectins and gums;
insoluble fibers included cellulose, hemicelbuloses, and lignin.
Methods
All isotope-tracer and insulin-clamp studies were conducted
in subjects after a 12-h overnight fast. A primed, constant infu-
sion of sterile, pyrogen-free [6,6-2H2]glucose (98% 2H, MSD
Isotopes, Montreal, Canada) was administered into an arm
vein with a calibrated syringe pump (Harvard Apparatus, Inc,
Millis, MA) to determine endogenous glucose production dur-
ing the basal state (0-1 50 mm) and during the euglycemic insu-
bin clamp (1 50-330 mm) as previously described (1 5, 16). Arte-
rialized venous blood was withdrawn every I 5 mm during the
last hour of both the basal and insulin-clamp periods for iso-
tope enrichment (16, 21). The 10 mU.m2.min insulin-
clamp technique was employed as previously described (13-
16). Three blood samples for glucose and insulin concentra-
tions were obtained before initiating the infusion of crystalline
porcine insulin (Velosubin, Nordisk USA, Bethesda, MD).
Four minutes after the start ofthe insulin infusion, glucose in
water (200 g/L) was administered at an initial empirical rate
with subsequent adjustments in the rate to maintain basal glu-
cose levels (1 3-16). Arterialized blood for estimation of plasma
glucose was obtained from a hand vein at 5-mm intervals; for
insulin concentrations blood was taken at 30-mm intervals
during the clamp. Plasma glucose was clamped during each
study at the plasma glucose concentration obtained just before
each individual’s first study. Euglycemia was maintained in all
subjects studied during the control, HCF diet, and post-HCF
diet periods with mean coefficients ofvariation for plasma glu-
cose of 3.9 ± 1 %. Subjects were awake and supine throughout
the studies, and the volume of blood removed and fluids in-
fused were similar for all subjects.
Plasma [6,6-2H2]glucose enrichment was measured as the
glucose-acetate-boronate derivative by combined gas chroma-
tography-mass spectrometry with selected ion monitoring (16,
21). Plasma glucose and insulin concentrations were deter-
mined as previously described (1 3, 1 5, 16). Serum cholesterol
by guest on July 10, 2011www.ajcn.orgDownloaded from
526 FUKAGAWA ET AL
TABLE 2
Blood variables and glucose disposal rates in young and old subjects before and after 4 wks ofan HCF diet*
Before After
Both Young Old Both Young Old
(n= 12) (n=6) (n=6) (n= 12) (n=6) (n6)
Fasting glucose
(mmol/L) 5.33 ±0.08 5.44 ±0.05 5.27 ±0.17 5.05 ±0.05t 5.05 ±0.05 5.05 ±0.1 lt
Fasting insulin
(pmol/L) 65.3 ±8.2 67.4 ±7.2 64.6 ±15.8 49.5 ±5.5t 50.2 ±7.2 50.9 ±lOOt
Serum cholesterol
(mmol/L) 5.66 ±0.30 5.17 ±0.18 6.15 ±0.52 4.40 ±O.43t 3.80 ±O.20t 4.99 ±O.49t
Serum triglyceride
(mmol/L) 1.21 ±0.26 0.79 ±0.07 1.62 ±0.46 1.46 ±0.40 1.32 ±0.19 1.60 ±0.80
Glucose disposal
(mol.kg’.min) 18.87± 1.70 21.15± 2.44 16.21 ±2.05 23.87± 2.78 27.81 ±5.00 19.60± 2.39
Clamp insulin
(pmol/L) 205.9 ±13.2 203.8 ±25.8 207.4 ±10.8 21 1.7 ±20.6 221.0 ±39.5 202.3 ±17.2
*j±5EM
tt Significantly different from before values: t <0.01, t <0.05, §p <0.02.
and triglyceride measurements were made on an SMA- 12 Au-
toanalyzer (Technicon Instruments Corp. Tarrytown, NY).
Basal-state rates of endogenous glucose production (Ra)
were calculated by the isotope-dilution equation (16, 2 1). Dur-
ing the euglycemic insulin clamp, glucose kinetics were deter-
mined by using Steele’s equations in their derivative form for
the nonsteady state (22). The evaluation of glucose turnover
based on the primed continuous infusion oflabeled glucose was
validated for both steady and nonsteady states (23). During the
insulin-clamp studies, glucose disposal rates were expressed as
the mean values observed during 15-20-mm periods. Steady-
state glucose disposal rates were calculated as the mean of the
last 30 mm ofthe study after correction for endogenous glucose
output. Glycosuria did not occur during these studies. Steady-
state insulin concentrations were calculated as a mean value of
samples obtained from 150 to 330 mm.
St atistical analysis
All data are expressed as mean ± SEM. Differences between
glucose disposal rates, fasting plasma glucose and insulin con-
centrations, and serum cholesterol and triglyceride concentra-
tions before and after the HCF diet were analyzed by analysis
of variance and paired ttests (BMDP Statistical Package, Los
Angeles). Values are based on data for young and older adults
combined except when indicated because preliminary analysis
revealed that there was no age effect with respect to changes in
glucose disposal and substrate and hormone concentration.
Results
There was <2% variation in mean body weight during the
entire study period (young subjects 78 ± 4 and 77 ± 6 kg and
elderly subjects 68 ± 5 and 67 ± 5 kg for periods 1 and 2, respec-
tively). As expected, energy intake was higher in the young than
in the elderly subjects (‘U- 3#{216}#{216}#{216}vs 1700 kcal/d, respectively).
Frequent interviews with the subjects did not suggest any im-
portant changes in the physical-activity patterns ofthe individ-
ual subjects during the study.
The effect of the HCF diet on cholesterol, triglyceride, glu-
cose, and insulin concentrations are shown in Table 2. Fasting
serum cholesterol was slightly higher in the old than the young
subjects (p <0.07). The HCF diet decreased the cholesterol
concentration by 27% and 19% (p <0.01) in the young and old
subjects, respectively; glucose by 5% (p <0.01); and insulin by
24% (p <0.01). Serum triglyceride concentrations were not
significantly different between the groups (p =0.39) and were
not affected by the HCF diet (p =0.34). These constituents of
blood returned to prestudy values after four of six young and
all ofthe elderly subjects resumed their free-choice diets for 2-
4 wk.
Basal hepatic glucose production was unchanged by the HCF
diet (11.66 ± .56 before and 11.10 ± .56 imol.kg’ .min’ af-
ter). Steady-state insulin concentrations during the 10 mU.
m2min eugbycemic clamps were 205.9 ± 13.2 before and
2 1 1.7 ± 20.6 pmol/L after the HCF diet and 2 16.0 ± 23.0
pmol/L in the four post-HCF-diet studies. Hepatic glucose out-
put during the insulin clamp periods was similar before and
after the HCF diet (3.33 ± 1 .1 1 vs 3.33 ± 1.66 mol .kg’.
min ‘, respectively).
Mean glucose disposal after 2 1-28 d on the HCF diet in-
creased by 143% of basal (Table 2). Four of the 12 subjects (2
young and 2 elderly) showed a slight decline in glucose disposal
after the diet. There was no independent age effect on the re-
sponse to the HCF diet. However, when the differences in glu-
cose disposal were examined in the two age groups separately,
the young subjects exhibited a statistically significant increase
(p <0.05) whereas the old subjects exhibited a more modest
increase (p <0. 10) (Table 2).
In the four young men studied after they had resumed their
self-selected diets for 2-4 wk, glucose infusion rates necessary
to maintain euglycemia were similar to those seen in the studies
before the initiation of the HCF diet (16.2 ± 3.90 vs 17.01
±1.52 mol.kg’ .min’).
Discussion
The results ofthese studies reveal that an HCF diet increases
significantly the sensitivity of peripheral tissues to physiologic
concentrations of insulin in healthy adults and thus demon-
by guest on July 10, 2011www.ajcn.orgDownloaded from
HCF DIET AND INSULIN SENSITIVITY 527
strate an extrahepatic effect of these diets. This response oc-
curred in the absence ofapparent changes in the control of he-
patic glucose output. In addition, as previously reported (1-4,
6-8), the HCF diet lowered fasting plasma glucose, insulin, and
cholesterol concentrations. This effect was reversed in <2 wk
after the pre-HCF ad libitum diet was resumed. Fasting serum
triglyceride concentrations were not significantly affected by
the diet.
These findings confirm the results ofa report by Nesteb et al
(24), who demonstrated that HCF diets do not alter hepatic
glucose production. However, these findings conflict with Nes-
tel et al’s findings that HCF diets do not have an effect on pe-
ripheral glucose uptake. Methodobogic differences probably ac-
count for this discrepancy. Nestel et al’s experimental dietary
period was 10 d compared with the 2 1-28 d used in this study.
Thus, it may be that a period of 10 d is too short to allow for
an effect of an HCF diet. The insulin rate used by Nestel et al
resulted in circulating insulin concentrations at the upper
range ofthe dose-response curve for insulin vs glucose uptake.
This would diminish the likelihood ofdetecting changes in in-
sulin sensitivity (13). The change noted by Nestel et ab, while
not statistically significant, was in the same direction that we
noted. In our studies steady-state insulin concentrations were
equivalent to those associated with glucose disposal rates of
slightly less than half of the maximum rate observed during
euglycemia ( I 3), thus increasing the likelihood ofdetecting sig-
nificant improvements in glucose disposal.
Although the present study does not permit analysis of rela-
tive contributions ofdietary fiber, complex carbohydrate, or fat
content in the HCF diet to changes in glucose homeostasis, it
does clarify a possible mechanism responsible for the improve-
ment in carbohydrate economy with such a diet. Because our
studies were conducted in subjects who were in the postabsorp-
tive state and receiving both glucose and insulin intravenously.
our approach is not confounded by immediate effects of the
HCF diet on the intestinal phase ofglucose utilization. It fob-
lows that the effect ofthe HCF diet extends beyond the gastro-
intestinal phase ofglucose uptake and must, therefore, involve
a change in the responsiveness ofperipheral tissues to insulin-
mediated glucose uptake. Our subjects did not show a change
in basal hepatic glucose output or a difference in its suppression
by insulin during the HCF diet period. Variability in the source
and amount of fiber and carbohydrate content in HCF diets.
length ofdietary period. subject selection. and differences be-
tween the metabolic states of the subjects make it difficult to
compare the present data with results from previous studies.
Although there is evidence that HCF diets produce splanch-
nic effects that improve carbohydrate homeostasis (7). there is
increasing evidence linking HCF diets to metabolic effects be-
yond the gut. Thus. in rats fed high-carbohydrate diets there
appears to be an increase in glycolytic enzyme activity in the
liver in addition to a higher insulin receptor number and
affinity (3), suggesting metabolic effects ofthe diet distant from
those occurring in the intestine. The present observation are
consistent with studies by Hjollund et al (25) in type II (non-
insulin-dependent) diabetics demonstrating increased insulin
sensitivity and receptor binding with I-ICF diets. These investi-
gators found an increased in vivo insulin sensitivity as assessed
by the intravenous insulin-tolerance tests and an increase in
insulin binding and glucose transport in isolated fat cells ob-
tamed from their patients. Riccardi et al (26) attempted to de-
termine the separate effects ofthe fiber and carbohydrate corn-
ponents of the diet on the control of glucose metabolism in
both type I (insulin-dependent) and type II diabetics, and these
investigators suggested that an increase of digestible carbohy-
drate without an increase in dietary fiber does not improve con-
trol ofblood glucose concentrations. However, their diets were
not as high in carbohydrate (53%) as that employed in the pres-
ent study and the dietary period was only 10 d. Similarly, other
investigators proposed that increased cereal fiber (27) or guar
gum (28) in the diet of type II diabetic patients improved pe-
ripheral insulin sensitivity. Recently, Chen et ab (29) reported
that 3-5 d ofan 85% carbohydrate diet improved insulin sensi-
tivity and glucose tolerance in elderly individuals as assessed
by the frequent sample intravenous glucose-tolerance test.
However, the effects of a longer dietary period or of fiber were
not examined. In the present study the change in glucose dis-
posal in response to the HCF diet in the elderly subjects was less
dramatic than seen in the young subjects. Because the subject
number was small, it is important to examine further the
differential effects of high carbohydrate and high fiber per se
on glucose tolerance in the aging population, emphasizing sub-
jects who have demonstrated impaired glucose tolerance on
screening.
Because of current interest in promoting HCF diets in the
general population (30, 3 1) and the present evidence of nonali-
mentary effects on glucose metabolism, further longer-term
studies are required, particularly to clarify the relative roles of
the available carbohydrate, dietary fiber, and lipid component
of such diets on glucose-insulin interrelationships in healthy
young and old individuals as well as in diabetic individ-
uals(32). U
We gratefully acknowledge the contribution ofEdwina Murray and
Helene Cyr and the technical assistance of Martha Duffy. Marilyn Jack-
son. Mary Frazier, Charles Couet. the staff of the MIT Clinical Re-
search Center, and the staffofthe Mass Spectrometry Unit at the Shri-
ncr’s Burns Institute, Boston. in the performance and analysis of these
studies.
References
I.Anderson JW. Chen WL. Plant Fiber. Carbohydrate and lipid me-
tabolism. Am J Clin Nutr 1979:32:346-63.
2. Villaume C. Beck B. Gariot P. Desalme A, Debry G. Long-term
evolution of the effect of bran injestion on meal-induced glucose
and insulin responses in healthy man. Am J Clin Nutr 1984:40:
I023-6.
3. Anderson JW. The role ofdietary carbohydrate and fiber in the
control ofdiabetes. Adv Intern Med 1980:26:67-96.
4. Kinmouth AL. Mangus RM. Jenkins PA. Smith MA. Baum JD.
Whole foods and increased dietary fiber improve blood glucose
control in diabetic children. Arch Dis Child 1982:57:187-94.
5. American Diabetes Association. Principlesofnutrition and dietary
recommendations for individuals with diabetes. Diabetes Care
I979: 2:520-3.
6. Anderson JW. Chen WL. Seiling B. Hypolipidemic effects of high-
carbohydrate. high-fiber diets. Metabolism I 980: 29:55 I -8.
7. Jenkins DJA. Jenkins AL. The clinical implications ofdietary 11-
ber. In: (Draper HH. ed.) Advances in Nutrition Research. Vol 6.
New York: Plenum Press. 1984:169-202.
8. Brunzell JD. Lerner RL. Hazzard WR. Porte D Jr. Bierman EL.
Improved glucose tolerance with high carbohydrate feeding in mild
diabetes. N EnglJ Med 197 1:284:521-4.
by guest on July 10, 2011www.ajcn.orgDownloaded from
528 FUKAGAWA ET AL
9. Kromhout D, Bosschieter EB, deLezenne Coubander C. Dietary
fiber and 10-year mortality from coronary heart disease, cancer,
andallcauses. Lancet l982;2:5l8-20.
10. Food and Nutrition Board. Toward healthful diets. Washington,
DC: National Academy Press, 1980.
1 1. Davidson D. The effect of aging on carbohydrate metabolism: a
review ofthe English literature and a practicalapproach to the diag-
nosis ofdiabetes mellitus in the elderly. Metabolism l979;28:688-
705.
12. Himsworth HP, Kerr RB. Age and insulin sensitivity. Clin Sci
l939-42;4: 153-7.
13. Rowe JW, Minaker KL, Pallotta JA, Flier JS. Characterization of
the insulin resistance ofaging. J Clin Invest 1983;7l:l58l-7.
14. Fink RI, Kolterman 0G. Kao M, Olefsky JM. Mechanisms of the
insulin resistence ofaging. J Clin Invest 1983;71:1523-35.
15. Meneilly GS, Minaker KL, Ebahi D, Rowe JW. Insulin action in
aging man: evidence for tissue-specific differences at low physio-
logic insulin levels. J Gerontol 1987;42: 196-201.
16. Fukagawa NK, Minaker KL, Rowe JW, Matthews DE, Bier DM,
Young VR. Glucose and amino acid metabolism in aging man:
differential effects ofinsulin. Metabolism 1988;37:37 I -7.
17. Metropolitan Life Insurance Company. New weight standards for
men and women. Stat Bull Metropol Life Insur Co l959;40:l-4.
18. National Diabetes Data Group. Classification and diagnosis of dia-
betes mellitus and other categories ofglucose intolerance. Diabetes
1979;28: 1039-57.
19. Anderson JW, Chen W, Ward K. Composition offoods commonly
used in diets for persons with diabetes. Diabetes Care 1978; 1:
293-6.
20. Anderson JW, Bridges SR. Tietyen J, Gustafson NJ. Dietary fiber
content ofa simulated American diet and selected research diets.
Am J Clin Nutr l989;49:352-7.
21. Bier DM, Leake RD. Haymond MW, et ab. Measurement of “true”
glucose production rates in infancy and childhood with 6,6-dideu-
tero-glucose. Diabetes l977;26: 10 16-23.
22. Steele R. Influence of glucose loading and of injected insulin on
hepatic glucose output. Ann NY Acad Sci 1959;82:420-30.
23. Radziuk J, Norwich KH, Vranic M. Experimental validation of
measurements ofglucose turnover in nonsteady state. Am J Phys-
iob l978;234:E84-93.
24. Nestel PJ, Nolan C, Vazelmans J, Cook R. Effects ofa high starch
diet with low or high fiber content on post-absorptive glucose utili-
zation and glucose production in normal subjects. Diabetes Care
1984;7:207-l0.
25. Hjollund E, Pederson 0. Richelsen B, Beck-Nielson H, Sorenson
NS. Increased insulin binding to adipocytes and monocytes and
increased insulin sensitivity of glucose transport and metabolism
in adipocytes from non-insulin dependent diabetics after low fat!
high-starch/high-fiber diet. Metabolism l983;32: 1067-75.
26. Riccardi G, Rivellese A, Pacioni D, Genovese 5, Mastranzo P.
Mancini M. Separate influence of dietary carbohydrate and fibre
on the metabolic control in diabetes. Diabetobogia l984;26:I 16-
21.
27. Karlstrom B, Vessby B, Asp NG, et al. Effects ofan increased con-
tent of cereal fiber in the diet of type II (non-insulin dependent)
diabetic patients. Diabetologia 1984;26:272-7.
28. Tagliaferro V, Cassader M, Bozzo C, et al. Moderate guar-gum ad-
dition to usual diet improves peripheral sensitivity to insulin and
lipaemic profile in NIDDM. Diabete Metab 1985; 1 1:380-5.
29. Chen M, Bergman RN, Porte D Jr. Insulin resistance and beta-
cell dysfunction ofaging: the importance ofdietary carbohydrate. J
Clin Endocrinol Metab l988;67:951-7.
30. US Department ofAgricubture, US Department ofHealth, Educa-
tion and Welfare. Dietary guidelines for Americans. Washington,
DC: US Government Printing Office, 1980.
31. Kushi LH, Lew RA, Stare FJ, et al. Diet and 20-year mortality
from coronary artery disease: the Ireland-Boston Diet-Heart
Study. N EnglJ Med 1985;3l2:81 1-8.
32. Reaven GM. How high the carbohydrate? Diabetobogia 1980; 19:
409-13.
by guest on July 10, 2011www.ajcn.orgDownloaded from
