B Gl
mp
Di
rudy
We roin
(SU f Afr
we ensit
IGT an id
Eig e GI
bo (PLA
kg ose t
glu for 2
mo eral
(DI m gl
mo seru
inc par
C-p 2 an
an AC g
sig se in
rem creas
mo LAC
fol o bas
in ITS
for peripheral insulin resistance (as assessed by DIs). GITS was well tolerated without any symptoms suggestive of either
hypoglycemia or significant weight gain. In summary, long-term chronic glipizde GITS administration improved glucose
ho
in h
glu
wi
hig
© 2
T
dys
mi
and
etio
is
fam
late
the
or
lati
T
F
He
S
S
00R
A
Sta
bus
©
0
d
414meostasis by increasing beta-cell responsiveness to glucose, improving Si, as well as significantly improved DI, but not Sg,
igh-risk, obese African Americans with IGT. Our study demonstrated that GITS appears to prime beta cells to intravenous
cose stimulation resulting in restoration of physiologic acute first- and second-phase insulin secretion in African Americans
th IGT. We conclude that GITS might be considered as a useful agent in the primary prevention of type 2 diabetes in
h-risk, obese African American patients with IGT.
004 Elsevier Inc. All rights reserved.
HE PREVALENCE of type 2 diabetes has increased in
several populations to epidemic proportions.1-5 Beta-cell
function and insulin resistance characterize the hyperglyce-
a found in patients with impaired glucose tolerance (IGT)
type 2 diabetes in several populations.6-11 The earliest
logic lesion in the development of IGT and type 2 diabetes
unknown, but it is presumed to be genetic with strong
ilial and environmental components. The disease has a long
ncy period with well-described precursors and predictors in
prediabetic phase. This provides an opportunity to introduce
implement diabetes prevention programs in high-risk popu-
ons.12-15
he prevalence of diabetes and its associated complications
is higher in minority populations residing in the United
States.3-5 This is particularly so for African Americans, who
have an extraordinary propensity for type 2 diabetes and the
related long-term complications when compared to white
Americans.3-5 We6,16 and others17 have previously demon-
strated that African Americans with and without IGT and type
2 diabetes manifest higher peripheral hyperinsulinemia and
greater insulin resistance when compared to their white coun-
terparts. Thus, nondiabetic African Americans with IGT could
be targeted for primary diabetes prevention. In this regard, the
Diabetes Prevention Program (DPP) demonstrated that lifestyle
modification (diet and exercise with the goal of losing 7% of
the baseline weight) and metformin reduced the incidence of
type 2 diabetes in patients with IGT by 58% and 31%, respec-
tively, in all US ethnic populations including African Ameri-
cans.12 Furthermore, Buchanan et al14 have reported that tro-
glitazone reduced the incidence of type 2 diabetes in previous
gestational Latino/Hispanic women, a population with tremen-
dous propensity for type 2 diabetes. However, oral sulfonylurea
(SU) agents were not included in the DPP, perhaps for the fear
of severe or fatal hypoglycemia. Thus, whether long-acting SU
could prevent or delay the development of type 2 diabetes in
high-risk African Americans remains uncertain.
rom the Ohio State University College of Medicine and Public
alth, Columbus, OH.
ubmitted January 10, 2003; accepted November 7, 2003.
upported by NIH NIDDK Grant No. DK 481287 and GCRC
R034.
ddress reprint requests to Kwame Osei, MD, FACE, FACP, Ohio
te University College of Medicine, 491 McCampbell Hall, Colum-
, OH 43210.
2004 Elsevier Inc. All rights reserved.eneficial Metabolic Effects of Chronic
With Impaired Glucose Tolerance: I
of Type 2
Kwame Osei, Scott Rhinesmith, T
examined the long-term metabolic effects of glipizide gast
), in impaired glucose-tolerant (IGT), first-degree relatives o
assessed glucose homeostasis, beta-cell function, insulin s
before and at yearly intervals for 24 months of GITS or
hteen IGT patients were randomized to receive either glipizid
dy mass index [BMI], 32.9  6.3 kg/m2) or identical placebo
/m2) for 24 months. Each of the subjects underwent oral gluc
cose tolerance test (FSIGT) at baseline and yearly intervals
del method. The ability of beta cell to compensate for periph
). Chronic administration of glipizide GITS attenuated seru
nths when compared to baseline (0 months). In contrast,
rease in the IGT/PLAC group at 12 and 24 months when com
eptide levels progressively increased in the GITS group at 1
d C-peptide responses remained unchanged in the IGT/PL
nificant improvement in the blunted acute first insulin relea
ained blunted in the IGT/PLAC group. We found that Si in
nths(P < .05) versus baseline, but deteriorated in the IGT/P
lowing GITS therapy at 12 and 24 months when compared t
the IGT/PLAC group throughout the study period. Chronic GT
ear
026-0495/04/5304-0017$30.00/0
oi:10.1016/j.metabol.2003.11.016ipizide in Obese African Americans
lications for Primary Prevention
abetes
Gaillard, and Dara Schuster
testinal therapeutic system (GITS), a potent sulfonylurea
ican American patients with type 2 diabetes. To this end,
ivity (Si), and glucose effectiveness (Sg) in patients with
entical placebo in a randomized, double-blind manner.
TS (GITS, 5 mg/d, n
9; mean age, 43.3 8.7 years; mean
C, n
9; mean age, 41.5  5.7 years; mean BMI, 39  4.2
olerance test (OGTT) and frequently sampled intravenous
years. Si and Sg were determined by Bergman’s minimal
insulin resistance was calculated as the disposition index
ucose responses to oral glucose challenge at 12 and 24
m glucose levels at fasting and during OGTT tended to
ed to baseline. Serum insulin (P < .05 to 0.01) and serum
d 24 months versus 0 months. In contrast, serum insulin
roup. During FSIGT, chronic GITS was associated with
the IGT patients at 12 and 24 months. These parameters
ed in the IGT/GITS group at 12 months (P < .01) and 24
group. Similarly, the DIs significantly (P < .01) increased
eline. In contrast, DI did not change from baseline values
partially restored the ability of beta cells to compensateherefore, we tested the hypothesis that SU could reverse the
ly beta-cell dysfunction and ultimately improve glucose
Metabolism, Vol 53, No 4 (April), 2004: pp 414-422

hom
To
gas
XL
cal
tien
in
Po
T
rela
ifes
we
deg
wit
or
crit
the
fas
afte
tha
pre
enc
live
end
sub
act
apa
bris
stre
at l
or c
by
the
sub
exp
Stu
A
clin
Bo
ligh
as w
ma
anc
circ
the
fere
Me
W
ins
infu
OG
E
in t
12-
ins
of
2-m
min
cos
Fre
F
the
era
for
the
blo
exo
ple
glu
sam
50%
min
olis
sec
8, 1
160
cen
and
Tw
A
sub
to
mo
C
M
N
bef
A
bod
dia
415GLIPIZIDE GITS AND GLUCOSE HOMEOSTASIS IN AFRICAN AMERICANS WITH IGTeostasis in high-risk African American patients with IGT.
this end, we investigated the metabolic effects of a glipizide
trointestinal therapeutic system (glipizide GITS; Glucotrol
, Pfizer Pharmaceutical Inc, New York, NY) and an identi-
placebo in first-degree relatives of African American pa-
ts with type 2 diabetes who had IGT treated for 24 months
a randomized, double-blind manner .
MATERIALS AND METHODS
pulations
he study group consisted of 18 subjects who were first-degree
tives of African American patients with type 2 diabetes and man-
ted IGT during an oral glucose tolerance test (OGTT). The subjects
re recruited during community screening for diabetes in the first-
ree relatives (offspring and siblings) of African American patients
h type 2 diabetes. The categories of glucose tolerance were defined
classified according to the World Health Organization (WHO)
eria,18 which was the classification at the time of initial screening of
subjects in 1996. Patients with IGT were defined as those with
ting serum glucose less than 140 mg/dL and 2-hour serum glucose
r a 75-g oral glucose challenge greater than 140 mg/dL, but less
n 199 mg/dL. We excluded individuals who had diabetes (newly or
viously diagnosed) who were taking medications known to influ-
e glucose and insulin metabolism. We also excluded subjects with
r, heart, lung, and kidney diseases and those who participated in
urance exercise or indulged in regular competitive sports. All of the
jects answered a simple questionnaire on physical activity. The
ivity level was described as (a) sedentary (no extra physical activity
rt from walking and activity of daily living), (b) moderate (tennis,
k walking, swimming, etc at least 3 times per week), and (c)
nuous (weight lifting, wrestling, racket ball, marathon, jogging, etc
east 3 times per week). Subjects who participated in an endurance
ompetitive sport were excluded. Informed written consent approved
the Institutional Review Board for Human Biomedical Research at
Ohio State University, Columbus, OH was obtained from each
ject after the risks entailed in the study have been thoroughly
lained.
dy Protocol
fter a 10- to 12-hour overnight fast, the subjects reported to the
ical research center of the Ohio State University Medical Center.
dy weight and height were measured with the subject wearing a very
t gown and without shoes. Body mass index (BMI) was calculated
eight (kilograms) divided by square of height (meters). Lean body
ss and body fat composition were measured by bioelectrical imped-
e analyzer.19 Body fat distribution was measured as the waist-to-hip
umference ratios. Waist circumference was measured at the level of
umbilicus (with the subject in standing position) and hip circum-
nce at the level of the greater trochanter (in the standing position).
tabolic Studies
ith the subject in the supine position, an intravenous needle was
erted into the forearm vein and kept patent with 0.9% normal saline
sion.
TT
ach subject was instructed to ingest at least 250 g of carbohydrate
heir regular meals for at least 3 days prior to the test. After a 10- to
hour overnight fast, blood samples were drawn for serum glucose,
ulin, and C-peptide at t
0 minutes. The subjects then ingested 75 g
oral glucose load (Glucola, Baltimore, MD; 250 mL) over a
Am
coninute period. Blood samples were drawn at t
30, 60, 90, and 120
utes for serum glucose, insulin, and C-peptide concentrations. Glu-
e tolerance status of the subjects was defined by the WHO criteria.18
quently Sampled Intravenous Glucose Tolerance
ollowing a 10- to 12-hour overnight fast, subjects were admitted to
research center for a frequently sampled intravenous glucose tol-
nce test (FISGT). Two intravenous needles were inserted into the
earm veins and kept patent with 0.9% normal saline infusion with
subject in supine position. One intravenous line was used to draw
od samples and the other to administer the intravenous glucose and
genous insulin as previously described.8,16,17,20,21 Four blood sam-
s were obtained at t
20,10,5, and 0 minutes for basal serum
cose, C-peptide, and insulin concentrations. The average of the 4
ples was taken as the basal level. Thereafter, 0.3 g/kg glucose as
dextrose water was infused over a 1-minute period: At t
19
utes, intravenous insulin (0.05 U/kg Humulin, Eli Lilly, Indianap-
, IN) dissolved in 30 mL of 0.9% normal saline was infused over 60
onds. Blood samples were obtained at frequent intervals at t
2, 5,
0, 12, 14, 16, 19, 22, 24, 25, 27, 30, 40, 60, 70, 90, 120, 140, 150,
, and 180 minutes for serum glucose, C-peptide, and insulin con-
trations. All samples were centrifuged at 4°C and the sera frozen
stored at 20°C until assayed.
enty-Four–Month Longitudinal (phase II) Follow-up Study
fter satisfying study entry requirements and baseline studies, the
jects were randomized in a double blind, placebo-controlled manner
receive either GITS (5 mg/d) or identical placebo (PLAC) for 24
nths. The clinical characteristics of the 2 subgroups of African
Table 1. Baseline Clinical and Metabolic Characteristics
of First-Degree Relatives of African American Patients
With Type 2 Diabetes With IGT
Parameter Placebo (n
9) GITS (n
9) P Value
linical characteristics
Age (yr) 41  4.7 43.3  8.7 NS
Body weight (kg) 108.9  18.8 95.8  16.3 NS
BMI (kg/m2) 39.0  4.2 32.9  6.3 .02
LBM (kg) 51.3  13.9 61.6  7.6 .01
BFM (%) 48.7  7.1 38.3  7.5 .01
WHR 0.93  0.37 0.88  0.07 NS
etabolic parameters
Glucose (mg/dL)
Fasting 83  8 88  12 NS
30 min 142  20 150  20 NS
2-h PP 160  20 153  14 NS
Insulin (
U/mL)
Fasting 15.5  6.4 13.4  6.5 NS
30 min 44.6  18.6 63.6  29.0 NS
2-h PP 96  38 94  58 NS
C-peptide (ng/mL)
Fasting 2.93  1.20 3.20  2.13 NS
30 min 5.60  2.24 7.07  2.82 NS
2-h PP 10.29  4.01 11.29  4.40 NS
OTE. Values are mean  SD. Metabolic parameters were obtained
ore and during standard oral glucose tolerance test.
bbreviations: BMI, body mass index; LBM, lean body mass; BFM,
y fat mass; WHR, waist-to-hip circumference ratio; PP, postpran-
l; NS, not significant.ericans with IGT are listed in Table 1. Group 1 (IGT/GITS, n
9)
sisted of subjects with IGT receiving GITS, and group 2 (IGT/