Sustained pectin ingestion: Effect on gastric emptying and glucose tolerance in non-insulin-dependent diabetic patients

Abstract

The effects of sustained pectin ingestion on gastric emptying, glucose tolerance, and hormone responses were studied in 12 stable, non-insulin-dependent (type 2) diabetic patients. Patients were placed on a 2400 kcal, low-fiber (3 g) diet for 2 wk, followed by 4 wk of an isocaloric diet supplemented with 20 g apple pectin/d. Gastric-emptying half-time, plasma glucose, glucagon, and human pancreatic polypeptide levels were determined. Gastric-emptying half-time was prolonged 43% (p less than 0.025) by pectin supplementation and returned to normal 3 d after its discontinuation. Fiber supplementation decreased the incremental area under the glucose tolerance curve from 34.8 +/- 3.0 to 27.9 +/- 3.2 mmol/L (p less than 0.01) but did not affect hormonal responses to a meal. Sustained pectin ingestion slowed the gastric-emptying rate and improved glucose tolerance; however, a direct relationship could not be demonstrated between changes in gastric emptying and changes in the incremental area under the glucose curve (r = 0.22).

Full text

Am J C/in Nuir 1988;48:1413-7. Printed in USA. © 1988 American Society for Clinical Nutrition 1413
Sustained pectin ingestion: effect on gastric emptying
and glucose tolerance in non-insulin-dependent
diabetic patients14
Sheldon E Schwartz, MD; Robert A Levine, MD; Ruth S Weinstock, MD, PhD; Susan Petokas, MSN;
ChrystaA Mills, BS, RDT; andFDeaver Thomas, MD
ABSTRACT The effects of sustained pectin ingestion on gastric emptying, glucose toler-
ance, and hormone responses were studied in 12 stable, non-insulin-dependent (type 2) dia-
betic patients. Patients were placed on a 2400 kcal, low-fiber (3 g) diet for 2 wk, followed by 4
wk of an isocaloric diet supplemented with 20 g apple pectin/d. Gastric-emptying half-time,
plasma glucose, glucagon, and human pancreatic polypeptide levels were determined. Gastric-
emptying half-time was prolonged 43% (p <0.025) by pectin supplementation and returned
to normal 3 d after its discontinuation. Fiber supplementation decreased the incremental area
under the glucose tolerance curve from 34.8 ± 3.0 to 27.9 ± 3.2 mmol/L (p <0.01) but did
not affect hormonal responses to a meal. Sustained pectin ingestion slowed the gastric-empty-
ing rate and improved glucose tolerance; however, a direct relationship could not be demon-
strated between changes in gastric emptying and changes in the incremental area under the
glucosecurve(r =0.22). AmJClin Nutr 1988;48:1413-7.
KEY WORDS Glucagon, gastrin, pancreatic polypeptide, adaptive intestinal mecha-
nisms, reversibility offiber effects
Introduction
The addition of nondigestible fiber to the diet im-
proves glucose tolerance in both insulin-treated and non-
insulin-treated diabetic patients (1-9). Our previous
study (10) demonstrated that sustained pectin supple-
mentation delayed gastric emptying in normal volun-
teers without affecting glucose tolerance. Jenkins et a!
(1 1) noted that there were no changes in the glucose re-
sponse to a carbohydrate load when pectin was added to
the diets ofnormal volunteers.
Pectin supplementation may enhance glucose toler-
ance in man by decreasing the rate of gastric emptying,
inhibiting intestinal glucose absorption, modifying the
hormonal response to a meal, or any combination of
these mechanisms (8, 10, 12-14). We attempted to deter-
mine whether the reported effect of pectin ingestion on
glucose tolerance in diabetic patients was related to de-
layed gastric emptying and/or to alterations in plasma
hormone responses.
Methods
Experimentaldesign and diets
The design of this study was similar to that of our previous
study (10) with normal adult volunteers except that in the pres-
ent study, the diabetic patients did not receive a-cellulose and
intestinal perfusion studies were not conducted.
Twelve non-insulin-dependent, non-ketosis-prone, adult-
onset (type 2) diabetic patients (seven men, five women; aged
37-66 y; duration of diabetes, 3-10 y) were evaluated in the
Clinical Research Center ofthe SUNY Health Science Center.
Five of the 12 patients were considered obese (> 85 kg). The
study was undertaken after approval was granted by the Institu-
tional Review Board for the Protection ofHuman Subjects. In-
formed written consent from all individuals was obtained be-
fore the study. All patients were selected from our diabetes
clinic, had been in reasonable metabolic control as determined
by their frequent fasting plasma glucose concentrations for 6
mo, and had required no alteration in their medications during
1From the Departments ofMedicine and Radiology(Nuclear Mcdi-
cine), State University of New York, SUNY Health Science Center at
Syracuse, Syracuse, NY.
2Presented in part at the 85th Meeting ofthe American Gastroenter-
ological Association in New Orleans on 22 May 1984.
3Supported in part by grant RR 229 from the General Clinical Re-
search Center Program of the Division of Research Resources, Na-
tional Institutes of Health.
4Address reprint requests to RA Levine, Department of Medicine,
Section of Gastroenterology, University Hospital, SUNY Health Sci-
ence Center, 750 East Adams Street, Syracuse, NY 13210.
Received September 8, 1987.
Accepted for publication January 27, 1988.
by guest on July 10, 2011www.ajcn.orgDownloaded from

1414 SCHWARTZ ET AL
the 6-mo period. Nine patients were maintained on neutral
protamine Hagedorn (NPH) insulin and three were using oral
hypoglycemic agents. Two patients were maintained on gly-
buride (one patient received 10 mg twice daily and the other
received 5 mg/d) and one patient received 500 mg tolbutamide
twice daily. Two patients had peripheral neuropathy as deter-
mined by neurological examination; no patients had clinically
significant gastroparesis, nephropathy, retinopathy, or auto-
nomic dysfunction as observed by postural hypotension or
beat-to-beat variation with deep breathing.
Adetailed dietary history of each subject was obtained; it
included an assessment ofthe height, weight, presence of food
allergies or intolerances, activity level, and daily average caloric
intake ofeach volunteer. Volunteers were then placed for 2 wk
on a 2400-kcal (50% carbohydrate, 30% fat, 20% protein), low-
fiber (3 g crude fiber) isocaloric diet. Bran and bran products,
whole-grain products, nuts, berries, and dried fruits were to-
tally excluded and fresh fruits and vegetables were partially re-
stricted. During the next 4 wk the volunteers ingested a high-
fiber diet that included an additional 20 g/d apple-pectin pow-
der (Solgar Inc, Lynbrook, NY). The fiber was baked into
muffins. One muffin (6.7 g pectin) was ingested with each meal.
The only difference between the diet used in our previous study
(10) and this diet was the incorporation of 5 mL of vanilla fla-
voring into each batch of 12 muffins to increase palatability.
The muffins were frozen to prevent spoilage and were heated
before use. The diets were equivalent in nutrients and energy
content and differed only in their fiber contents. Diets were de-
signed to maintain patients’ weight throughout the study. Iso-
caloric diets were based on standard values (1 5). The low-fiber
diet was reinstituted for as long as 3 wk after the completion of
the 4-wk, high-fiber phase.
All patients were cooperative and maintained a strict caloric
intake (mean 2350 kcal/d, range 1820-2740 kcal/d) as deter-
mined by dietary history. Fiber supplements were well toler-
ated. Compliance was monitored by two ofthe authors (SP and
CM) in weekly interviews. Compliance was considered good if
90% of the muffins were ingested and if the containers were
returned empty.
At the conclusion ofeach dietary period (the 2-wk, low-fiber
diet; the 4-wk, high-fiber diet; or 3 d, 1 wk, or 3 wk after re-
sumption of the low-fiber diet), subjects who fasted overnight
(16 h) ingested a low-fiber, low-carbohydrate breakfast meal
surface-labeled with 99”technetium-sulfur colloid (mTc).
Eighteen and one-half millibecquerels of colloid was injected
into each of two eggs through the shell; the eggs were subse-
quently cooked until hard-boiled. A tenacious bond is formed
when the isotope is cooked in whole eggs (16). The eggs were
mashed, mixed with 30 mL of mayonnaise, and made into a
sandwich. Canned grape juice (236 mL) with 1 5mL of added
sugar was ingested with the sandwich. The total meal mass was
43 1 g. The meal contained 690 kcal: 75 mg carbohydrate (59%)
as 50 g simple sugars and 25 g complex carbohydrates, 75 g
protein(l4%), and 18 g fat (27%).
Gastric-emptying studies
Evaluation ofthe rate ofgastric emptying was performed by
our previously described method (10) for studying solid meal
components except that the isotope was mixed with eggs in-
stead ofwith tuna fish. The in vivo stability ofthe 99mTc4abeled
omelet sandwich meal was previously demonstrated ( 16).
After ingestion of the radiolabeled meal, subjects were
placed in a supine position with the scintillation detector (Ohio
Nuclear, Subsidiary of Technicare, Inc, Solon, OH) centered
over the stomach. Preset, timed (5 mm), static images of the
stomach were taken at 5, 15, 30, 60, 90, and 120 mm with a
standard-field ‘y-scintillation camera (Ohio Nuclear Inc, Solon,
OH) interfaced with a computer (Digital Equipment Corp,
Marlboro, MA). Gastric emptying was expressed as the disap-
pearance half-time ofthe radiolabeled solid meal (t#{189})in mm-
utes. The percentage ofradioactivity remaining in the stomach
30 mm after ingestion of the test meal was subtracted from
100% and expressed as the percentage of the meal emptied in
30 mm. Gastric-emptying studies were performed on all pa-
tients after they completed the 2-wk, low-fiber and the 4-wk,
high-fiber dietary phases. Gastric-emptying studies were also
performed 3 d, 1 wk, and 3 wk on seven, eight, and two pa-
tients, respectively, after they resumed the low-fiber diet.
Blood glucose response to mealfteding and hormonal studies
The morning insulin dose was held until completion of the
blood glucose response to meal feeding. After ingestion of the
test meal and concurrently with gastric-emptying studies, glu-
cose concentrations were determined as previously described
(10) for the low-fiber and pectin-supplemented-fiber diets but
not during the recovery low-fiber diets. Radioimmunoassays
for gastrin, glucagon, and human pancreatic polypeptide
(HPP) were performed by previously reported techniques (10).
Plasma samples were frozen and subsequently analyzed for an-
tibodies to HPP by Drlan L Taylor(University of California-
Los Angeles School of Medicine).
Statistical analysis
Data are expressed as means ± SEM. Statistical comparisons
for significant differences were made by paired Student’s ttests
for single comparisons (gastric emptying and glucose-tolerance
tests) and analysis ofvariance (plasma-hormonal comparisons)
(17). The t#{189}was calculated by linear-regression analysis by the
method ofleast squares (17).
Results
Effect ofpectin supplementation on gastric-emptying
rates
Gastric-emptying times were prolonged from 83.4
±6.8 mm before pectin administration to 1 19.3 ± 14.4
mm after supplementation (p <0.025) as shown in Fig-
ure 1.In 3 of the 12 patients maintained on the pectin-
supplemented diet, all of whom were receiving insulin
and were in metabolic control, gastric-emptying times
were hastened (Fig 1). After patients resumed the low-
fiber diet, gastric-emptying times reverted toward base
line in all of the eight patients whose gastric-emptying
times were substantially delayed after pectin supplemen-
tation. In those subjects restudied 3 d, 1 wk, and 3 wk
after they resumed the low-fiber diet, gastric-emptying
times decreased 26, 37, and 45%, respectively, when
compared with each oftheir individual pectin-diet values
(Fig 1).
The mean gastric-emptying times of the breakfast
meal were within the normal range. This was established
from data obtained from 1 3 normal, nonobese control
subjects on a low-fiber diet (10).
by guest on July 10, 2011www.ajcn.orgDownloaded from

LOW PECTIN
FIBER DIET
DIET X4wk
LOW
FIBER
DIET
x3days
LOW
FiBER
DIET
xIwk
LOW
FIBER
LET
x3wk
C
E
04
Li
I-.
0
z
>-
I-
a-
Li
()
I-
U,
0
DIET
7.8
6.7
5.6
4.4
3.3
2.2
1.1
p
LL
LU
2O
40
-1--J
Z(/)
w
WO-
0
z
PECTIN INGESTION AND DIABETES 1415
FIG 1. Effect ofsustained pectin supplementation on gastric-empty-
ing time in 12 diabetic patients who discontinued the pectin-supple-
mented diet and resumed the low-fiber diet.
Effect ofpectin supplementation on glucose homeostasis
and hormonal studies
Fasting plasma glucose concentration was 1 1 .5 ±1.2
mmol/L before low-fiber supplementation and dietary
control, 6. 1 ± 1 .7 mmol/L after the 2-wk low-fiber phase,
and 12.0 ± 1 .6 mmol/L after the 4-wk pectin-supple-
mentation period. The incremental area under the curve
was 34.8 ±3.0 mmol/L after the low-fiber diet compared
with 27.9 ±3.2 mmol/L after fiber supplementation (p
<0.01, Fig 2). Patients’ mean weights before and after
fiber supplementation were unchanged, 78.2 ± 2 and
78.3 ± 3 kg, respectively. There was no difference be-
tween the incremental area under the curve in blood glu-
cose response to meals obtained 4 wk after pectin supple-
mentation compared with 3 d and 1 wk after resumption
of the low-fiber diet. There was no correlation between
the changes in the gastric-emptying rate and the changes
in the incremental area under the curve induced by fiber
supplementation (r =0.22).
Pectin supplementation did not alter any of the
plasma glucagon, gastrin, and HPP responses measured
after ingestion of the test meal (Table 1). None of
the patients’ base-line plasma samples had antibodies
to HPP.
Side effects
Four patients described a feeling of fullness or in-
creased flatulence during fiber supplementation. No pa-
tients developed altered bowel habits. One subject devel-
oped recurrent clinical symptoms (sweating and head-
ache) of mild hypoglycemia which was confirmed by
diminished blood glucose. This occurred at end of the
second week ofpectin supplementation. This patient was
taking 90 U daily of NPH insulin. His dosage of insulin
was decreased for 2 d and he was then placed on his origi-
nal dose ofinsulin. He had no other clinical episodes that
suggested hypoglycemia. No other patients had alter-
ation oftheir insulin or oral hypoglycemic dosage.
Discussion
We previously reported that sustained pectin ingestion
delayed gastric emptying in healthy volunteers without
altering glucose tolerance or hormonal responses to a
low-fiber meal (10). We have now shown that sustained
pectin ingestion delays gastric emptying and alters the
blood-glucose response to meal feeding in non-insulin-
dependent (type 2) diabetic patients. This is one of two
studies that demonstrate that long-term dietary fiber sup-
plementation (without increasing carbohydrate content)
improves glucose homeostasis in diabetic patients (18).
Our study design is unique in that we used a low-fiber
0
LOW-FIBER DIET .-.
PECTIN-SUPPLEMENTED #{163}- #{163}
0 30 60 90 120 150 180
TIME (mm)
FIG 2. Effect ofsustained pectin supplementation on the incremen-
tal plasma glucose concentration change from fasting levels after the
test meal in 12 diabetic patients. ±SEM values at 30, 60, 90, 120, 150,
and 180 mm for the low-fiber and pectin-supplemented diets were, re-
spectively, 0.83 and 1.00, 0.89 and 1.05, 0.78 and 0.89, 0.94 and 1.05,
0.83 and 0.89, and 1.00 and 1.1 1 mmol/L.
by guest on July 10, 2011www.ajcn.orgDownloaded from

1416 SCHWARTZ ET AL
TABLE 1
Effect ofpectin on plasma hormones (pmol/L)
Pectin-
Time Low-fiber
diet supplemented
diet 3 d
postfiber 1 wk
postfiber
Glucagon
0123± 6 124± 8 111± 7 131±16
30 119± 5142±11 125± 5 122± 7
60 125± 6 133±12 133± 7 126± 12
90 117± 4 133±10 127± 6 120± 7
120 126± 7 139±10 120± 5120± 4
150 125± 7 136±11 137±18 128± 9
180 115± 6 123±10 124± 5124± 6
Gastrin
0 20±6 22±6 16±2 31±11
30 23±6 23±8 24±6 48±13
60 25±5 27±8 29±9 46±13
90 22± 4 30± 8 29± 8 44±11
120 20± 3 29±10 28± 8 40±10
150 19± 2 22± 4 27± 8 40±12
180 17± 2 20± 4 22± 6 39± 12
Human pancreatic
polypeptide
014± 3 28±15 14± 1 32± 16
30 62±6 51±2 72±2 65±40
60 78±16 82± 2 101± 2 124±27
90 70± 582±16 78± 5 88±29
120 55±11 65± 3 96± 2 72± 16
150 32± 16 51 ±3 72±20 69±14
180 35±12 62±10 65±19 73±25
Sj; SEM. Time in mm.
breakfast meal when we measured gastric-emptying
rates. In other studies (3, 1 3, 19), fiber was provided with
test meals in the same amounts that were used daily dur-
ing the treatment phase. Our methodology identifies
adaptive changes and differentiates them from acute
phenomena. Ours is also the first study to demonstrate
the reversibility of fiber-induced delay in gastric-empty-
ing rates in diabetic patients. The adaptive effects of sus-
tamed pectin ingestion on gastric emptying were revers-
ible within 3 d.
It is unclear why pectin supplementation induces a de-
lay in the rate ofgastric emptying and improves glucose
tolerance. Sustained pectin supplementation did not
affect glucose concentrations after a test meal although it
prolonged gastric emptying in normal volunteers (10).
Several studies (5, 1 3, 20, 2 1) suggested that the im-
proved glucose homeostasis noted in diabetic patients
may involve mechanisms other than pectin-induced pro-
longation of gastric emptying or alterations in hormone
responses to a meal and that delayed gastric emptying
may not be responsible for improved metabolic re-
sponses. Our study supports this contention because no
correlation existed between delayed emptying and re-
duction in the incremental area under the glucose curve.
Normal volunteers and diabetic patients developed
similar delays in their rates ofgastric emptying while they
received pectin supplementation yet the ingestion of pec-
tin only affected the glucose responses of the diabetics.
Effects offiber ingestion on glycemic responses in diabet-
ics may be exaggerated and related to their known fl-cell
dysfunction. Our data support the concept that sustained
fiber ingestion improves glucose tolerance in non-insu-
lin-dependent diabetics (1-3, 5, 7-9). Pure-glucose loads
increase postprandial hyperglycemia to a greater extent
than do mixed-carbohydrate meals (22). Physical factors
in food may determine the glycemic response of nondia-
betics and diabetics; gastric inhibitory polypeptide may
be a hormone that varies with changing physical factors
offood (23). It is conceivable that normal subjects might
have had glycemic alteration with a different experimen-
tal design (5, 1 1). Our study did not examine the possibil-
ity of an effect of pectin supplementation on peripheral
insulin sensitivity in type 2 diabetics.
We found that the rates ofgastric emptying ofthe dia-
betic patients were within the normal values ofour labo-
ratory for nonobese, healthy control subjects, a finding
previously confirmed by other studies (18, 24, 25) of dia-
betic subjects who did not have autonomic neuropathy.
The delays in emptying were variable. The pectin-in-
duced delay in gastric emptying appeared to have no
effect on the shape ofthe postprandial glucose curve (Fig
2), which is similar to the observations ofRay et al (18).
Although pectin ingested with a meal may delay the
delivery of gastric contents into the jejunum by altering
gastric motor activity or hormonal responses, our data
derived from the ingestion of a low-fiber meal support
the concept that adaptive intestinal mechanisms may
underlie improved glucose homeostasis. We conclude
that pectin ingestion improves glucose tolerance by un-
known mechanisms that may be involved with but can-
not be fully explained by delayed gastric emptying.
Areported problem with the use ofdietary pectin sup-
plementation is its low palatability and the frequency of
associated gastrointestinal side effects (18). Our studies
demonstrated that vanilla-flavored pectin muffins in-
gested three times daily are well tolerated in a diabetic
population for 4 wk. Supplementation with nondigest-
ible fiber such as pectin should not be avoided in diabetic
subjects. Although we have demonstrated that sustained
pectin supplementation produces a reversible adaptive
effect on gastric emptying, we were unable to show sim-
ilar reversibility ofthe improved glucose tolerance within
the week following the resumption of the low-fiber diet.
The fact that fiber improved postprandial glycemia be-
yond its effect on gastric-emptying times does not neces-
sarily mean that prolonged gastric emptying was causally
unrelated to improvement in glucose homeostasis. How-
ever, it is likely that sustained improvement of diabetic
control by an enriched fiber diet would require continu-
ous meal supplementation with fiber (26). 13
References
1. Anderson JW, Chen WJL. Plant fiber carbohydrate and lipid me-
tabolism. Am J Clin Nutr 1979; 32:346-63.
2. Simpson HCR, Lousley 5, Geekie M, et al. A high carbohydrate
leguminous fibre diet improves all aspects ofdiabetic control. Lan-
cat 198l;l:1-5.
3. Anderson JW, Midgley WR, Wedman B. Fiber and diabetes. Dia-
betes Care 1979;2:369-79.
by guest on July 10, 2011www.ajcn.orgDownloaded from

PECTIN INGESTION AND DIABETES 1417
4. Parsons SR. Effects ofhigh fiber breakfasts on glucose metabolism
in noninsulin-dependent diabetics. Am J Clin Nutr l984;40:66-
71.
5. Jenkins DJA, Jenkins AL. Dietary fiber and the glycemic response.
Proc Soc Exp Biol Med 1985; 180:422-3 1.
6. Vogelberg KH, Kubler HGW. Gastric emptying and blood glucose
in insulin-dependent diabetes mellitus and autonomic neuropathy.
Deutsche Med Wochen 1985; 110:1249-53.
7. Kiehm TG, Anderson JW, Ward K. Beneficial effects of a high
carbohydrate, high fiber diet on hyperglycemic diabetic men. Am
JClin Nutr 1976; 29:895-9.
8. Jenkins DJA, GoffDV, Leeds AR, et al. Unabsorbable carbohy-
drates and diabetes: decreased postprandial hyperglycemia. Lancet
l976;2:172-4.
9. Miranda PM, Horwitz DL. High-fiber diets in the treatment of dia-
betes mellitus. Ann Intern Med 1978; 88:482-6.
10. Schwartz SE, Levine RA, Singh A, ScheideckerJR. Sustained pec-
tin ingestion delays gastric emptying. Gastroenterology 1982;83:
812-7.
1 1. Jenkins DJA, Leeds AR, Houstin H, et al. Carbohydrate tolerance
in man after six weeks of pectin administration. Proc Nutr Soc
l977;36:60A(abstr).
12. Jenkins DJA, Gassull MA, Leeds AR, et a!. Effect ofdietary fiber
on complications ofgastric surgery: prevention ofpostprandial hy-
poglycemia by pectin. Gastroenterology 1977;73:215-7.
13. Holt 5, Heading RC, Carter DC, et al. Effect ofgel fiber on gastric
emptying and absorption of glucose and paracetamol. Lancet
1979; 1:636-9.
14. Jenkins DJA, Wolever TMS, Leeds AR, et al. Dietary fibers, fiber
analogues, and glucose tolerance: importance ofviscosity. Br Med
J l978;l:l392-4.
15. Pennington JAT, Church HN, eds. Bowes and Church’s food val-
ues ofportions commonly used. 14th ed. Philadelphia: JB Lippin-
cottCo, 1985.
16. Knight LC, Malmud LS. Tc-99m-Ovalbumin labelled eggs; corn-
parison with other solid food markers in vitro. J Nucl Med
198 1;22:P38(abstr).
17. Murphy EA. Biostatistics in medicine. Baltimore: Johns Hopkins
University Press, 1982.
18. Ray TK, Mansell KM. Knight LC, Malmud LS, Owen OE, Boden
G. Long-term effects ofdietary fiber on glucose tolerance and gas-
thc emptying in noninsulin-dependent diabetic patients. Am J
Gin Nutr 1983; 37:376-81.
19. Jenkins DJA. Dietary fiber and carbohydrate metabolism. Diabe-
tea Care l979;2: 175-92.
20. Torsdottir I, Alpsten M, Anderson D, Brummer RiM, Anderson
H. Effect ofdifferent starchy foods in composite meals on gastric
emptying rate and glucose metabolism. I Comparisons between
potatoes, rice and white beans. Hum Nutr Gin Nutr 1984;38C:
329-38.
21. Lawaetz 0, Blackburn AM, Bloom SR, Artas Y, Ralphs DNL.
Effect ofpectin on gastric emptying and gut hormone release in the
dumping syndrome. Scand J Gastroenterol 1983; 18:327-36.
22. Reaven GM. Effects ofdifferences in amount and kind of dietary
carbohydrate on plasma glucose and insulin responses in man. Am
J Clin Nutr 1979; 32:2568-78.
23. Simpson RW, McDonald J, Wahlqvist ML, Alley L, Outch K.
Food physical factors have different metabolic effects in nondiabet-
ics and diabetics. Am J Clin Nutr 1985;42:462-9.
24. Hongo M, Lin YF, Satake K, Toyata T, Okuyama 5, Goto Y. Gas-
tric emptying in patients with diabetes mellitus with and without
autonomic neuropathy. Gastroenterology 1982; 90: 1464(abstr).
25. Scarpello JHB, Barber DC, Hague RV, et al. Gastric emptying of
solid meals in diabetics. Br Med J l976;2:6l-3.
26. Chenon D, Phaka M, Monnier LH, Colette C, Osettia A, Mirouze
J. Effects ofdietary fiber on postprandial glycemic profiles in dia-
betic patients submitted to continuous programmed insulin infu-
sion. Am J Clin Nutr 1984;40:58-65.
by guest on July 10, 2011www.ajcn.orgDownloaded from