Diabetes Research and Clinical Practice 53 (2001) 137-139
The effect of Extend Bar® containing uncooked cornstarch on night-time
glycemic excursion in suspects with type 2 diabetes
Maryellen Dyer-Parziale *
Casa Pacifica Medical Clinic, 1722 S Lewis Rd, Camarillo, CA 93012, USA
Received 23 November 1999; received in revised form 24 April 2000; accepted 15 May 2000
The objective of this study was to determine the effects of ingesting a snack bar containing uncooked
cornstarch (Extend Bar®, Clinical Products, Limited, Key Biscayne, FL) on nocturnal glycemic excursion
in 28 adults (ages 22-78 years) with type 2 diabetes mellitus (mean HbAlc 8.21 ± 1.28%). Thirteen subjects
were treated with oral agents, eight with a combination of insulin and oral agents, and seven with insulin
alone. Subjects ingested the study bar (Extend Bar®, containing 30 g of total carbohydrate, including 5 g of
uncooked cornstarch, 3 g protein, and 3 g fat) for three evenings followed by a placebo bar for 3 evenings
(30 g of total carbohydrate, 3 g protein, and 3 g fat), or vice versa. Pre-snack before bedtime, midnight and
before breakfast finger stick blood glucose levels were compared to determine the incidence of
hypoglycemia ( < 60 mg/dl), hyperglycemia ( > 250 mg/dl), and to calculate any differences in the group's
mean blood glucose levels when ingesting the study versus the placebo bar. There were no episodes of
hypoglycemia or hyperglycemia The mean blood glucose levels pre-snack at bedtime were nearly identical
(Extend Bar® value 117.5 ± 45.6 mg/dl; placebo bar value 117.3 ± 40.0 mg/dl; P = 0.977), and lower at
midnight and before breakfast on the Extend Bar® nights compared to the placebo bar nights (Extend Bar®,
midnight value 127.9 ± 3 1.0 mg/dl; placebo bar midnight value 148.2 + 32.1 mg/dl; P = 0.0001; Extend
Bar® breakfast value 114.2 ± 15.8 mg/dl; placebo bar breakfast value 158.49 ± 30.3 mg/dl; P < 0.0001).
These data suggest that ingesting Extend Bar® containing uncooked cornstarch as the nighttime snack may
be an effective strategy to lesson the frequency of nocturnal and morning hyperglycemia in subjects with
type 2 diabetes. © 2001 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Extend Bar®; Cornstarch; Hyperglycemia; Hypoglycemia; Type 2 diabetes
Controlling glycemic excursion in subjects with type 2 diabetes mellitus is an
important goal to limit the devastating long-term complications of this disease [1,2].
One strategy, which has been shown to effectively limit glycemic excursion in
patients with type 1 diabetes has been the use of a snack bar containing uncooked
cornstarch ([3-5]) - a complex carbohydrate, which is slowly absorbed from the
gastrointestinal track over a 6-9 h period [6,7]. Kaufman et al. showed a decrease in
the incidence of hypoglycemia at midnight and in the morning after the ingestion of a
snack bar containing uncooked cornstarch compared to a placebo bar with the same
grams of carbohydrate ([3-5]). Of equal importance, particularly in the management
of hyperglycemia prone patients with type 2 diabetes, there was no increase in
observed hyperglycemia after ingesting the cornstarch snack bar.
The present study was undertaken to determine if a snack bar (Extend Bar®, Clinical
Products, Limited, Key Biscayne, FL) containing 5 g of uncooked cornstarch could
decrease the incidence of hyperglycemia at midnight, 2-3 h post-ingestion, and the
next morning, 8-9 h post-ingestion, in adults with type 2 diabetes. Subjects were
chosen for participation regardless of whether they were receiving oral hypoglycemic
agents, insulin or combination therapy.
2. Materials and methods
Twenty-eight subjects (11 men and 17 women) with type 2 diabetes followed in Casa
Pacifica Medical Clinic in Ventura, CA were chosen for this study after informed
consent was obtained. The mean age of the subjects was 61.9 ± 14.5 years (range 23
months-78 years), the mean diabetes duration was 49.7 ± 53.6 months (range 2
months-20 years) and the mean HbA1c was 8.27 ± 1.3% (range 6.7 - 10.81/0). There
were 13 subjects treated with oral agents, 8 with insulin and 7 with combination
therapy. They routinely had a pre-bedtime snack.
The 28 study subjects with type 2
diabetes were randomly assigned to
ingest Extend Bar® for 3 evenings
followed by a placebo bar for three
evenings, or vice versa. The content of
the two 160 kcal bars are given in
Table 1. The subjects were blinded as
to the type of bar they ingested and
were instructed to follow their regular
diabetes routine during the study
period, including having a usual dinner
on all study nights. Pre-snack before
bedtime, midnight and before breakfast
finger stick blood glucose levels were
obtained with the patient's home glucose meter and results were recorded by the
subject on a data sheet. Results from these determinations were compared to
determine the incidence of hypoglycemia (number of glucose readings < 60 mg/dl),
hyperglycemia (number of glucose readings > 250 mg/dl), and to calculate any
differences in the group's mean blood glucose levels at the 3 evaluation times when
ingesting the cornstarch snack bar versus the placebo bar. Statistical analysis was
done with student's t-test.
There were no episodes of hypoglycemia or hyperglycemia pre-snack before bedtime,
at midnight or before breakfast during the study period. As shown in Fig. 1, the mean
blood glucose levels pre-snack at bedtime were nearly identical. However, the mean
blood glucose levels were significantly lower at midnight and before breakfast on the
nights that the Extend Bar® was ingested
compared to the placebo bar. There were
no significant differences in blood glucose
levels by type of diabetes treatment
regimen. The bars were well tolerated with
no episodes of gastrointestinal side effects.
These data suggest that adult subjects with
type 2 diabetes had significant reduction in
midnight and fasting blood glucose levels
on the nights that the Extend Bar®,
containing uncooked cornstarch, was
ingested compared to a placebo bar of equal carbohydrate, protein and fat content but
without uncooked cornstarch. This occurred regardless of the treatment regimen
followed, with subjects on oral agents, insulin and both showing reduction in fasting
and midnight blood glucose levels. It appears that ingestion of Extend Bar® as the
nighttime snack before bed may be an effective strategy to lesson the frequency of
nocturnal and morning hyperglycemia in subjects with type 2 diabetes without
increasing the risk of hypoglycemic events.
Uncooked cornstarch is a complex carbohydrate composed of 27% of the linear chain
dextrose polymer amylose and 73% of the branched chain dextrose polymer
amylopectin . Uncooked cornstarch is converted into maltose and other very small
glucose polymers in the small intestine by the hydrolytic effect of pancreatic amylase.
Intestinal epithelial enzymes allow for further hydrolysis and the slow absorption that
occurs into the portal circulation over at least a 6 h period . As a result of this slow
hydrolysis and absorption, and perhaps as a result of mixing the carbohydrate with fat
and protein , glucose entry into the blood stream may be delayed even further,
extending over a 6 h period. This helps avoid extremes of glycemic excursion and
reduces hyperglycemic and hypoglycemic episodes.
The study showed that ingesting Extend Bar® allowed for a significant reduction in
fasting blood glucose levels. Reduction in fasting glycemia is the goal of the intensive
management of type 2 diabetes. As outlined in the algorithms developed by the
UKPDS , the goal of therapy for type 2 diabetes is to achieve a fasting glucose
level between 110-120 mg/dL. That goal was achieved in this study for the subjects
when they ingested Extend Bar®. This would suggest that Extend Bar® should be a
useful adjunct to the diabetes regimen for subjects with type 2 diabetes on insulin,
oral agents, or both wishing to eat a snack before bedtime.
 The Diabetes Control and Complication Trial Research Group, The effect of intensive diabetes management on the development and
progression of long-term complications in insulin dependent diabetes mellitus. The diabetes control and complication trial, N. Engl. J.
Med. 329 (1993) 977-987.
 UKPDS Group, Lancet 352 (1998) 837.
 F.R. Kaufman, S. Devgan, Use of uncooked cornstarch to avert nocturnal hypoglycemia in children and adolescents with type I
diabetes. J. Diabetes Complications 10 (1996) 84-87.
 F.R. Kaufman, M. Halvorson, N.D. Kaufman, A randomized, blinded trial of uncooked cornstarch to diminish nocturnal
hypoglycemia at diabetes camp, Diabetes Res. Clin. Pract. 30 (1995) 205-209.
 F.R. Kaufman, M. Halvorson, N.D. Kaufman, Evaluation of a snack bar containing uncooked cornstarch in. subjects with diabetes,
Diabetes Res. Clin. Pract. 35 (1997) 27-33.
 Y.-T. Chen, M. Cornblath, J.B. Sidbury, Cornstarch therapy in type I glycogen-storage disease, N. Engl. J. Med. 310 (1984) 1721-
 R. Lozano, S.A. Chalew, A.A. Kowarski, Cornstarch ingestion after oral glucose loading: Effect on glucose concentrations, hormone
response, and symptoms in patients withpostprandial hypoglycemia syndromes, Am. J. Clin. Nutr. 52 (1990) 667-670.