Effect of metformin on patients with impaired glucose

tolerance
C. L. Li, C. Y. Pan, J. M. Lu, Y. Zhu, J. H. Wang, X. X. Deng, F. C. Xia, H. Z. Wang and
H. Y. Wang
Abstract
Aims To evaluate the effect of metformin on glucose metabolism, insulin sensitivity
and rate of conversion diabetes in people with impaired glucose tolerance (IGT).
Methods Seventy subjects with IGT were randomized under double-blind
conditions to receive either placebo (n= 37) or metformin (n =33) at a dosage of
250mg three times daily for a duration of 12 months. Glycaemic control, plasma
insulin and other biochemical indexes were assessed before and after 3, 6 and
12 months.
Results At 12months the conversion rate to diabetes was 16.2% in the placebo
group compared to 3.0% for the metformin group (P=0.011). Of subjects treated
with metformin for 12 months, 84.9% became normoglycaemic compared to
51.4% of those receiving the placebo. Signicant improvements in fasting glucose,
glucose tolerance and insulin sensitivity were found at 12 months and at
intermediate clinic assessments.
Conclusions Metformin can improve glucose metabolism in IGT patients and may
be a treatment option in their management of IGT subjects.
Diabet. Med. 16, 477481 (1999)
Keywords glycaemic control, impaired glucose tolerance, insulin sensitivity,
metformin
Abbreviations AUC, areas under the curve; BMI, body mass index; DM, diabetes
mellitus; IGT, impaired glucose tolerance; IRI, insulin resistance index; OGTT, oral
glucose tolerance test; UAE, urinary albumin excretion rate; WHR, waist-to-hip ratio
Introduction
The prevalence of impaired glucose tolerance (IGT) is
associated with geographical location, ethnicity and lifestyle
[1]. In developing countries, like China, there is strong
evidence that prevalence rates are rising, due in part to
improvements in socio-economic conditions and living
standards and an increasing life expectancy [2]. Private
enterprise, leading to more disposable income, and the
availability of Western style cuisine are factors contributing
to increased obesity and associated risk of glucose intolerance
and diabetes. As IGT patients have an increased risk of
cardiovascular disease [35] and are at risk from developing
frankdiabetes mellitus [68], there is growing recognitionthat
the active management of IGT subjects should be a key
objective in preventative medicine.
The biguanide metformin has been used in the treatment of
Type 2 diabetes mellitus (DM) for 40years [9,10] but few
studies have examined its role in impaired glucose tolerance
[11]. Metforminis anoral hypoglycaemic agent withanaction
on fasting and postprandial plasma glucose. It rarely causes
clinical hypoglycaemia [12].
Several studies have demonstratedaneffect of metforminon
insulin sensitivity, both in people with diabetes and in patients
at risk of developing diabetes [13]. We report the outcome of a
placebo-controlled study of metformin in IGT subjects, set up
to measure the conversion rate to frank diabetes and the effect
of treatment on obesity and lipids, both major risk factors for
cardiovascular disease.
R
Department of Endocrinology, Chinese PLA
General Hospital, Beijing, China
Received 14 August 1998; revised 11 January
1999; accepted 15 January 1999
Correspondence to: Dr Chunlin Li, Department of Endocrinology, Chinese PLA
General Hospital, 28 Fuxing Road, Beijing 100853, China.
E-mail: Licl@plagh.com.cn
1999 British Diabetic Association. Diabetic Medicine, 16, 477481 477
Subjects and methods
Study design
FromMay 1992toFebruary 1993, we undertooka population
screen of all the employees over the age of 30, currently
working in the 32 units of the Shougang Corporation, a heavy
industry enterprise based in Beijing. These data have been
reported [2]. Twenty-nine thousand, nine-hundred and thirty-
eight subjects (85% of total employees) were investigated
using a 75-g oral glucose tolerance test (OGTT). People with
pre-existing diabetes were excluded. We found a prevalence of
IGT of 4.19%, the total number of people with IGT being
1165. In 1994, these people with IGTwere re-examined using
75-g OGTT. Those whose results met the WHO criteria
(1985) for IGT in both 1992 and 1994 were eligible for our
study. Males and females aged 3060 years were eligible.
Patients with a history of ischaemic heart disease or renal or
hepatic disorders were excluded, as were patients who had
previously been treated with metformin.
People meeting the entry criteria were randomized under
double-blind conditions to receive either placebo or metfor-
min at a dosage of 250 mg three times daily for a duration of
12 months. The placebo was provided by the manufacturer of
metformin, Beijing Tian-An United Pharmaceutical Co. Ltd.
(Tokyo, Japan). After inclusion, follow-up visits were
scheduled at 2-monthly intervals. At each clinic visit, clinical
and biochemical investigations were undertaken and subjects
completed a single questionnaire to determine compliance and
nature any of side-effects of the medication.
The study was approved by the ethical committee for
pharmcological trials of the Chinese PLA Hospital and all
subjects gave formal consent.
Clinical and biochemical investigations
Weight, heightandwaist-to-hipratio(WHR) wererecordedfor
all subjects. The waist was measured at the point of maximum
girth between the subcostal margin and the anterior superior
iliac spine. Hip girth was recorded at the site of the greater
trochanter. Bodymassindex(BMI; weight/heightinkg/m
2
)was
calculated. After30minutes' rest, bloodpressurewasmeasured
twice in the same armby sphygmomanometer and the mean of
twomeasurements recorded. The restingECGwas recorded. A
venous blood sample was withdrawn for measurements of
fasting plasma glucose, HbA
1c
, insulin, cholesterol, triglycer-
ide, HDL cholesterol, GPT, BUNand creatinine. Twelve-hour
overnight urine samples were collected for the measurement of
urinary albumin excretion rate (UAE). The insulin resistance
index (IRI) was calculated using the homeostasis model
assessment (HOMA) method [14], where
IRI insulin=22:5e
lnglucose
All subjects were given75g anhydrous glucose and1and2h
blood samples withdrawn for the measurement of glucose and
insulin levels. The areas under the curve (AUC) were
calculated for glucose and insulin. These measurements were
undertaken at baseline andat 3, 6 and12 months of treatment.
Patient education in diabetes was provided every 3months.
Plasma glucose concentration were analysed by the glucose-
oxidase method using an autoanalyser (Model 75 Hitachi,
Tokyo, Japan) at the laboratory of the Chinese PLA General
Hospital. HbA
1c
was measured by the microcyclinder afnity
chromatograthy method with an intra-assay coefcient of
variation (CV) of 3.3% and interassay CV 4.7%. Insulin
concentrations were measured using the radioimmunoassay
method (American DPC Corporation, Tianjing, China) with
an intra-assay CV 6.2% and interassay CV 11.9%. UAE
concentration was measured using the radioimmunoassay
method (Atomic Energy Institute of China, Beijing, China)
with an intra-assay CV of 3.1% and interassay CV 4.6%.
Statistical analysis
Valuesareexpressedasmean6SDfornormallydistributeddata
or geometrical means withcondencelimits (insulinandUAE).
Differences betweenthe variables after metforminandplacebo
treatment wereanalysis bymultivariateanalysis of variancefor
repeat data. The statistical analysis was made on a computer
using SAS software package (SAS Institute, Cary, NC).
Results
Atotal of 90subjects were enrolledin the study but 20patients
were excluded from the analysis, leaving 70 evaluable
subjects. Twelve subjects were excluded from the metformin
group for the following reasons: tablet noncompliance, seven;
loss to follow-up, three; and gastrointestinal side-effects, two.
Eight subjects were excluded from the placebo group as
follows: tablet noncompliance, ve; loss to follow-up, two;
and raised liver enzymes, one. In the metformin group, there
remained 24 males and nine females, aged 49.0 6 1.3 years,
while in the placebo group, there were 26 males and 11
females, aged 50.0 6 1.1 years. Glucose tolerance data at
baseline for these subjects are given in Table 1. The groups
were well matched with no signicant differences. Patients
were 50 years old on average, with near normal weight and fat
distribution. The number with a positive family history for
Type 2 DM was also equal in the two groups.
On completion of 12months' treatment, 28 of the
metformin-treated patients reverted to normal glucose toler-
ance (84.9%), compared to 19 (51.4%) for placebo patients
(P=0.011). Six patients on placebo converted to frank
diabetes (16.2%) and this compared to one patient (3.0%)
on metformin (P= 0.011, Table 2).
The pattern of response over the 12 months is shown in
Table 1. At 3months, fasting plasma glucose was decreasedin
patients on metformin compared to placebo (P<0.05).
Signicant falls in fasting glucose were reported at 6 and
L
478 Effect of metformin on patients with IGT C. L. Li et al.
1999 British Diabetic Association. Diabetic Medicine, 16, 477481
R
T
a
b
l
e
1
P
a
t
t
e
r
n
o
f
c
l
i
n
i
c
a
l
a
n
d
b
i
o
c
h
e
m
i
c
a
l
c
h
a
n
g
e
s
d
u
r
i
n
g
1
2
m
o
n
t
h
s
o
f
t
r
e
a
t
m
e
n
t
P
l
a
c
e
b
o
g
r
o
u
p
(
n
=
3
7
)
M
e
t
f
o
r
m
i
n
g
r
o
u
p
(
n
=
3
3
)
0
3
6
1
2
m
o
n
t
h
s
0
3
6
1
2
m
o
n
t
h
s
F
P
F
a
s
t
i
n
g
g
l
u
c
o
s
e
(
m
m
o
l
/
l
)
7
.
3
6
1
.
0
6
.
9
6
1
.
1
7
.
2
6
1
.
4
6
.
2
6
1
.
3
6
.
9
6
0
.
9
6
.
2
6
0
.
8
*
6
.
2
6
0
.
8
*
*
*
5
.
0
6
1
.
1
*
*
2
1
.
7
9
0
.
0
0
0
1
1
-
h
p
l
a
s
m
a
g
l
u
c
o
s
e
(
m
m
o
l
/
l
)
1
1
.
7
6
2
.
5
1
2
.
2
6
2
.
9
1
2
.
5
6
3
.
1
1
2
.
2
6
3
.
2
1
1
.
7
6
2
.
7
1
1
.
6
6
2
.
2
1
0
.
9
6
2
.
2
*
9
.
9
6
2
.
7
*
*
3
.
4
2
0
.
0
2
4
9
2
-
h
p
l
a
s
m
a
g
l
u
c
o
s
e
(
m
m
o
l
/
l
)
9
.
0
6
1
.
0
9
.
2
6
2
.
3
8
.
0
6
2
.
5
7
.
4
6
2
.
5
9
.
1
6
0
.
9
8
.
5
6
1
.
9
7
.
8
6
1
.
8
6
.
0
6
2
.
0
*
1
8
.
5
1
0
.
0
0
0
1
A
U
C
g
l
u
c
o
s
e
1
9
.
9
6
2
.
3
2
0
.
2
6
4
.
1
2
0
.
1
6
4
.
5
1
9
.
0
6
4
.
5
1
9
.
7
6
3
.
0
1
9
.
0
6
3
.
0
1
7
.
8
6
3
.
0
*
1
5
.
4
6
3
.
7
*
*
8
.
5
7
0
.
0
0
0
1
H
b
A
1
c
(
%
)
7
.
4
6
0
.
8
7
.
4
6
1
.
2
6
.
3
6
1
.
1
7
.
3
6
0
.
9
7
.
3
6
0
.
8
7
.
2
6
1
.
7
5
.
8
6
0
.
8
6
.
6
6
1
.
3
*
1
8
.
0
7
0
.
0
0
0
1
F
a
s
t
i
n
g
i
n
s
u
l
i
n
(
m
U
/
l
)
4
.
9
6
2
.
0
6
.
0
6
1
.
9
6
.
9
6
1
.
5
6
.
6
6
1
.
6
4
.
6
6
1
.
8
3
.
7
6
1
.
5
*
*
4
.
9
6
1
.
4
*
*
4
.
5
6
1
.
7
*
2
.
0
0
0
.
1
2
8
9
1
-
h
i
n
s
u
l
i
n
(
m
U
/
l
)
2
4
.
0
6
1
.
9
3
2
.
4
6
2
.
1
3
7
.
2
6
1
.
7
4
0
.
7
6
1
.
6
2
6
.
3
6
2
.
1
2
4
.
0
6
1
.
9
2
3
.
4
6
1
.
8
*
*
2
4
.
6
6
1
.
5
*
*
*
1
.
6
1
0
.
2
0
3
1
2
-
h
i
n
s
u
l
i
n
(
m
U
/
l
)
2
2
.
4
6
1
.
9
2
3
.
4
6
2
.
0
2
6
.
3
6
1
.
9
3
2
.
4
6
1
.
9
2
4
.
6
6
1
.
9
2
1
.
4
6
2
.
0
1
9
.
1
6
1
.
9
1
9
.
1
6
1
.
8
*
*
2
.
9
3
0
.
4
3
6
2
A
U
C
i
n
s
u
l
i
n
3
9
.
8
6
1
.
7
4
9
.
0
6
2
.
1
5
6
.
2
6
1
.
7
6
0
.
3
6
1
.
6
4
2
.
7
6
1
.
9
3
8
.
0
6
1
.
8
3
6
.
3
6
1
.
7
*
3
8
.
0
6
1
.
4
*
*
*
1
.
4
0
0
.
2
5
6
0
I
R
I
2
.
1
3
6
2
.
1
2
.
1
6
6
1
.
4
2
.
5
2
6
1
.
6
2
.
0
2
6
1
.
4
1
.
6
9
6
1
.
2
1
.
1
1
6
0
.
5
*
*
1
.
4
9
6
0
.
6
*
*
1
.
1
8
6
0
.
5
*
*
3
.
2
0
0
.
0
3
0
0
U
A
E
(
m
g
/
m
i
n
)
8
.
1
3
6
2
.
5
7
.
7
6
6
2
.
1
8
.
5
1
6
2
.
5
8
.
5
1
6
2
.
9
5
.
8
9
6
2
.
0
5
.
7
5
6
1
.
9
6
.
0
3
6
1
.
8
5
.
0
1
6
1
.
9
*
0
.
6
8
0
.
5
6
7
0
C
h
o
l
e
s
t
e
r
o
l
(
m
m
o
l
/
l
)
5
.
2
2
6
1
.
0
4
.
7
9
6
0
.
8
4
.
2
1
6
0
.
4
5
.
0
2
6
0
.
7
5
.
2
7
6
1
.
2
4
.
8
7
6
1
.
0
4
.
2
0
6
0
.
8
4
.
8
9
6
0
.
9
2
9
.
0
9
0
.
0
0
0
1
T
r
i
g
l
y
c
e
r
i
d
e
s
(
m
m
o
l
/
l
)
1
.
7
9
6
1
.
2
1
.
7
8
6
1
.
1
1
.
5
7
6
1
.
0
1
.
3
8
6
0
.
6
1
.
9
3
6
1
.
4
1
.
8
9
6
2
.
6
1
.
3
8
6
0
.
6
1
.
6
3
6
1
.
4
2
.
8
5
0
.
0
4
9
6
S
y
s
t
o
l
i
c
B
P
(
m
m
H
g
)
1
3
2
6
2
0
1
3
0
6
1
8
1
2
8
6
1
6
1
2
9
6
2
1
1
3
4
6
1
8
1
3
1
6
1
6
1
2
8
6
1
4
1
2
4
6
2
0
4
.
9
8
0
.
0
0
3
6
D
i
a
s
t
o
l
i
c
B
P
(
m
m
H
g
)
8
8
6
1
1
8
5
6
1
1
8
4
6
1
0
8
4
6
1
3
8
9
6
1
2
8
5
6
1
1
8
4
6
1
0
8
4
6
1
3
7
.
1
1
0
.
0
0
0
3
W
H
R
0
.
8
7
8
6
0
.
0
7
0
.
8
8
0
6
0
.
0
7
0
.
8
8
1
6
0
.
0
7
0
.
8
7
3
6
0
.
0
7
0
.
8
5
3
6
0
.
0
5
0
.
8
5
1
6
0
.
0
5
*
0
.
8
5
4
6
0
.
0
6
0
.
8
4
0
6
0
.
0
6
*
1
.
2
1
0
.
3
1
3
9
B
M
I
(
k
g
/
m
2
)
2
6
.
0
6
2
.
3
2
6
.
7
6
2
.
4
2
6
.
1
6
2
.
3
2
6
.
4
6
2
.
4
2
6
.
4
6
2
.
4
2
5
.
9
6
2
.
5
2
5
.
3
6
2
.
4
*
2
5
.
0
6
2
.
2
*
6
.
4
0
0
.
0
0
0
7
M
e
t
f
o
r
m
i
n
v
s
.
p
l
a
c
e
b
o
*
P
<
0
.
0
5
*
*
P
<
0
.
0
1
*
*
*
P
<
0
.
0
0
1
I
n
s
u
l
i
n
a
n
d
U
A
E
:
g
e
o
m
e
t
r
i
c
a
l
m
e
a
n
s
w
i
t
h
c
o
n

d
e
n
c
e
l
i
m
i
t
s
.
O
t
h
e
r
s
d
a
t
a
:
m
e
a
n
6
S
D
.
Original articles 479
1999 British Diabetic Association. Diabetic Medicine, 16, 477481
12 months, respectively, and by the end of the study, the mean
fasting glucose was 5mmol/l for the metformin group and
6.2 mmol/l for the placebo group (P< 0.01).
An improved response to the glucose tolerance test was
observed for metformin treated subjects, with AUCvalues for
glucose being markedly lower than in the placebo group
(P<0.01). Changes in fasting glycaemia and glucose tolerance
were accompanied by a net fall of 0.7% for HbA
1c
in the
metformin group (6.6 61.3% vs. 7.3 60.9%, P= 0.0001 vs.
placebo).
Fasting insulin increased by over 20% in the placebo group
after 3 months and the elevated levels were maintained for the
remainder of the study. In contrast, insulin levels fell in the
metformin group, leading to signicant differences from the
placebo group over the 12-month duration of the study.
Changes in the glucoseinsulin ratio gave rise to signicant
improvements in the insulin sensitivity index over 12months.
Signicant reductions were noted at 12 months in the
metformin group for the waist-to-hip ratio (WHR) and BMI
(P<0.05).
Urinary albumin excretion was within the normal range
(< 12 mg/min) for subjects in both groups but fell at 12 months
from 5.89 to 5.01 mg/min (P<0.05) in the group treated with
metformin.
No signicant changes were observed for blood pressure,
cholesterol or triglyceride during the course of the study.
Compliance was good in the evaluable patients completing the
study. Three patients reported mild diarrhoea and nausea
when starting metformin, but the symptoms resolved with
long-term treatment. Six patients reported mild nausea on
placebo but the symptoms remitted with duration of
treatment. One placebo-treated patient developed raised liver
enzymes and required treatment.
On an intention-to-treat basis, excluding only ve patients
lost to follow-up, 32 of the metformin treated subjects became
normally glucose tolerant (76.2%) compared to 23 (53.5%)
for placebo patients. Six patients on placebo converted to
frank diabetes (14.0%) and this compared to three patients
(7.1%) on metformin, P= 0.091, Table 3.
Discussion
According to the WHO, in 1993 the prevalence of IGTwas 3
10% and the conversion rate of IGT to diabetes is about one-
third every 510 years. Our investigations of workers in the
Beijing Shougang corporation in 1992 showed a prevalence
rate of 4.19% for IGT in subjects with an age range of 30
60 years, in agreement with other studies in China [3].
In a previous study, we observed the natural history of 346
people with IGT. After a 2-year follow-up, 62 (17.9%) turned
out to be frankly diabetic, a conversion rate to diabetes of
8.95%per year [8]. Inour current study, the conversion rate to
diabetes inour placebo groupwas 16.2%per year, higher than
our previous observation. The small number of subjects in the
present study may contribute to this, rapid improvement in
socio-economic conditions and living standards is another
factor. The relatively high fasting plasma glucose levels are
also an important reason. However, we think the present high
conversion rate to diabetes in our placebo group may reect a
genuine trend of increasing DM prevalence and high conver-
sion rate from IGT to DM.
We have demonstrateda higher conversionrate fromIGTto
normal glucose tolerance after 12months of metformin-
treatment (84.9%) than in a placebo group (53.4%). Both
groupsreceivededucationconcerningdiet, exerciseandhealthy
lifestyle. Every 3 months, the education included basic knowl-
edge of diabetes, such as what diabetes is, the risk factors for
diabetes, the symptoms of diabetes and how to prevent it.
Subjects were alsogivenabasic knowledge of diet andexercise,
healthylifestyles, theeffects andside-effects of metformin, why
weuseit, andcommondrugs usedinIGTandDM. Wehavenot
formallyevaluatedthiseducationbut thedropinfastingplasma
glucose between 6 months and 12 months in both the placebo
andmetformingroups mayhave beenassociatedwithachange
of the lifestyle and decreased BMI.
The failure to see a signicant effect with an intention-to-
treat analysis may have two explanations. First, in the
metformin group, although 42 subjects were re-examined at
12 months, two IGT subjects had not taken metformin for
L
NGT IGT DM
n n % n % n % c
2
p
Metformin 33 28 84.9 4 12.1 1 3.0
Placebo 37 19 51.4 12 32.4 6 16.2 9.04 0.011
Table 2 The glycaemic status of patients at
12 months in 70 IGT patients
NGT IGT DM
n n % n % n % c
2
p
Metformin 42 32 76.2 7 16.7 3 7.1
Placebo 43 23 53.5 14 32.6 6 14.0 4.795 0.091
Table 3 The glycaemic status of patients at
12 months in 85 IGT patients
480 Effect of metformin on patients with IGT C. L. Li et al.
1999 British Diabetic Association. Diabetic Medicine, 16, 477481
nearly 11 months because of gastrointestinal side-effects, and
seven subjects with nontablet compliance took metformin for
less than 6 months. It must also be acknowledged that the
number observed was relatively small, andeasily inuencedby
small interference factors.
There is agrowingrecognitionthat patients withIGTshould
bemoreaggressivelytreated. Not onlydothesepatients havean
increasedriskof developingdiabetes but manyhave features of
the insulin resistance syndrome such as obesity, hypertension,
dyslipidaemiaandcoronaryheart disease. Metforminmight be
consideredasasuitabletreatment choiceasit hasbeenshownto
have benecial effects on insulin sensitivity, glucose and lipid
metabolismand body weight.
The present study suggests that metformin at low dose
(250 mg three times a day) is well tolerated and signicantly
improves glycaemic status and insulin sensitivity over
12 months, with an accompanying fall in body weight and
WHR. These ndings complement the work of Scheen et al.
[15] who showed that insulin sensitivity was improved in
android obese IGT subjects treated with metformin.
The improvement in body weight and WHR with metfor-
min has been shownby others bothin people with andwithout
[13,16] diabetes, but not in all cases [17]. Both diabetes
education and dietary advice may have contributed to the
improvements in BMI in our study, as a daily dose of 750mg
per day is considerably less thanthe average dose of metformin
for Type 2 diabetes (range 13g) used outside China. The
reduction in urinary albumin excretion is unexpected but has
been shown by others [18,19] after 612 months treatment
with metformin. The effect could be secondary to the
reduction in hyperglycaemia.
In our study, we diagnosed IGT using 1985 WHO criteria.
According to the newADAcriteria, 16 patients (48.5%) in the
metformin group and 20 patients (54.1%) in the placebo
group at baseline (c
2
0.217, P>0.05) would have been
considered diabetic on the basis of a fasting plasma glucose
of over 7 months. By the new criteria, we were treating some
early diabetes in our study.
We believe that this is the rst study todemonstrate aneffect
of metformin in reducing the conversion rate of IGT to
diabetes mellitus. Furthermore, the results conrm that
metformin is able to restore normal glucose tolerance in a
high percentage of Chinese patients with IGT (84.9%). Our
results suggest that metformin is of potential interest in the
management of IGTanddeserves more extensive evaluationin
longer termpreventative studies of diabetes mellitus. Both the
Early Diabetes Intervention Trial (EDIT) in the United
Kingdom and the Diabetes Prevention Programme 2 (DPP2)
in the United States with a minimumof 3 years' follow-up [20]
will help towards a future evaluation of drug therapy.
Acknowledgements
We wish to thank Dr H. Howlett for helpful assistance in the
preparation of this article.
References
1 King H, Rewers M. WHO Ad Hoc Diabetes Reporting Group:
Global estimates for prevalence of diabetes mellitus and impaired
glucose tolerance in adults. Diabetes Care 1993; 16: 157177.
2 Pan CY, Lu JM, Tian H, Kong XT, Lu XP, Yao C et al. Study of the
prevalence of diabetes mellitus in adults in the Shougang corporation
in Beijing. Diabetic Med 1996; 113: 663668.
3 Pan XR, Hu YH. Impaired glucose tolerance and its relationship to
ECG-indicated coronary heart disease and risk factors among
Chinese Da Qing IGT and diabetes study. Diabetes Care 1993; 16:
150156.
4 Jarrett RJ. Associations beween cardiovascular disease and impaired
glucose tolerance. IDF Bulletin 1996; 41: 810.
5 Jarrett RJ, McCartney P, Keen H. The Bedford survey: ten year
mortality rates in newly diagnosed diabetics, borderline diabetics
and normal glycaemic controls and risk indices for coronary heart
disease in borderline diabetics. Diabetologia 1982; 22: 7984.
6 Keen H, Jarrett RJ, McCartney P. The ten year follow up of the
Bedford survey (196272): glucose tolerance and diabetes. Diabe-
tologia 1982; 22: 7378.
7 Ramachandran A, Snehakatha C, Nalk RAS, Mohan V, Shobana R,
Viswanthan M. Signicance of impaired glucose tolerance in an
Asian Indian population: a follow-up study. Diabetes Res Clin Pract
1986; 2: 173178.
8 Chunlin L, Juming L, Changyu P et al. Two-year follow-up for
patients with impaired glucose tolerance. Chin J Endocrinol Metab
1997; 13: 104107.
9 Bailey CJ. Biguanides and NIDDM. Diabetes Care 1992; 15: 755
772.
10 Campbell IW, Howlett HCS. Worldwide experience of metformin as
an effective glucose lowering agent: a meta-analysis. Diabetes Metab
Rev 1995; 11: S57S62.
11 Papoz L, Job O, Eschwege E, Aboulker S, Cubeau J, Pequignot G,
et al. Effect of oral hypoglycemic drugs on glucose tolerance and
insulin secretion in borderline diabetic patients. Diabetologia 1978;
15: 373380.
12 Bailey CJ, Turner RC. Metformin. N Engl J Med 1996; 334: 574
579.
13 Hermann LS, Melander A. Biguanides: basic aspects and clinical
uses. In: Alberti KGMM, Defronzo RA, Keen H, Zimmet P, eds.
International Textbook of Diabetes Mellitus, 2nd edn. Chichester,
UK: John Wiley & Sons Ltd, 1997: 841864.
14 Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF,
Turner RC. Homeostasis model assessment: insulin resistance and B-
cell function fromfasting plasma glucose and insulin concentrations
in man. Diabetologia 1985; 28: 412419.
15 Scheen AJ, Letiexhe MR, Lefebvre PJ. Effect of metformin in obese
patients with impaired glucose tolerance. Diabetes Metab Rev 1995;
11: S69S80.
16 Fontbonne A, Charles MA, Juhan-Vague I, BIGPRO Study Group.
The effect of metformin on the metabolic abnormalities associated
with upper body fat distribution. Diabetes Care 1996; 19: 920926.
17 Hermann LS, Schersten B, Bitzen PO, Kjellstro m T, Lindga rde F,
Melander A. Therapeutic comparison of metformin and sulphony-
lurea, alone and in various combinations. Adouble-blind controlled
study. Diabetes Care 1994; 7: 11001109.
18 Campbell IW, Menzies DG, Chalmers J, Mobain AM, Brown ZRF.
One year comparative trial of metformin and glipizide in type 2
diabetes mellitus. Diabete Metab 1994; 21: 394400.
19 Nagi DK, Yudkin JS. Effects of metformin on insulin resistance, risk
factors for cardiovascular disease, and plasminogen activator
inhibitor in NIDDM subjects. Diabetes Care 1993; 16: 621629.
20 Chiasson JL. Possible therapeutic approaches to impaired glucose
tolerance. IDF Bull 1996; 41: 1620.
R
Original articles 481
1999 British Diabetic Association. Diabetic Medicine, 16, 477481