endogenous insulin secretion in newly diagnosed diabetic patients in saudi arabia
TRANSCRIPT
Diabetes Research and Clinical Practice, 8 (1990) 51-60 Elsevier 51
DIABET 00336
Endogenous insulin secretion in newly diagnosed diabetic patients in Saudi Arabia
0. S. Al-Attas ‘, M. A. Laajam2, M. S. Khan3 and A. Z. Al-Dress3
I Department of Biochemistry, College of Science, ‘Department of Medicine. College of Medicine, King Saud University and 3 Riyadh
Central Hospital, Riyadh. Saudi Arabia
(Received 30 November 1988) (Revision received 30 June 1989)
(Accepted 20 July 1989)
Summary
Diabetes mellitus is a major health problem in Saudi Arabia. The evaluation of endogenous insulin secretion at diagnosis has not yet been studied in this population. We have therefore studied fasting and post-glucagon stimulation levels of glucose, insulin and C-peptide in 216 newly diagnosed untreated diabetic patients. The mean k SD fasting insulin and C-peptide levels were 14.0 & 1.8 pU/ml and 1.8 + 0.4 ng/ml, while post-glucagon stimulation levels were 21.1 & 3 pU/ml and 2.4 + 0.4 ng/ml. There were significant post-stimulatory increment levels for insulin, from 4.9 to 13.7 pU/ml, and C-peptide from 0.2 to 1.3 ng/ml (P < 0.001). Such increments did not affect specified age distribution. We found a significant correlation between the fasting levels and post-stimulation levels of C-peptide and insulin. Obesity correlated with higher basal and post-stimulation levels of both hormones (r = 0.67, P -c 0.001). The mean f SD fasting insulin and C-peptide levels were 18.5 + 9.1 pUU/ml and 2.4 + 0.8 ng/ml for obese patients and 11.5 ? 5.1 pU/ml and 1.9 ? 1.1 ng/ml for non-obese patients. The type of diabetes among the Saudi adult diabetic patients studied is characterized by high basal C-peptide and insulin levels which increase significantly with stimulation, suggesting diminished but present endogenous B-cell function.
Key words: Saudi diabetics ; Glucagon stimulation test; Insulin ; C-peptide
Introduction
recognized in the recent classification
(classified as IDDM or type 1). Patients with late onset of the disease are generally not prone to spontaneous
classified as non-insulin-dependent
Publishers B.V. (Biomedical
52
teristically show profound insulin deficiency, and type 2 patients are known to comprise a hetero- geneous group with varying degrees of B-cell function - as in the recent study of Caucasian NIDDM patients and of an Indian population in Madras [ 31. A distinct subgroup of type 2 patients has been characterized showing an early require- ment for insulin to correct hyperglycemia [4]. Moreover, type 2 diabetes can be present during youth though this is uncommon [ 5,6]. With this heterogeneity it can be difficult to evaluate the disease and pursue appropriate treatment. Par- ticular difficulties arise in developing countries and non-Caucasian ethnic groups, where different patterns of diabetes are beginning to emerge [ 21.
The development of the C-peptide assay has provided an additional tool for direct studies of qualitative and quantitative residual B-cell function in diabetic patients [7,12].
The prevalence of diabetes mellitus has been studied in urban and rural Saudi Arabia con- current .with the rapid development of medical care [ 13,141, showing a high level of diabetes (more than 4%) among the population. Little is known of pancreatic function in these patients. The present study was therefore aimed at ascertaining residual B-cell function in newly diagnosed Saudi diabetic patients.
Patients and methods
Diabetic patients Two hundred and sixteen newly diagnosed adult Saudi diabetic patients (diagnosis made within 6 months of development of symptoms) presenting for the first time at the Central Hospital and King Abdulaziz University Hospital, Riyadh, Saudi Arabia, were studied. In this group 41.2% were male and 58.5% were female subjects, and the mean age was 46.9 + 12.9 years (SD) (range 13-95). Body mass index (BMI)ranged from 16.2 to 51.3 (mean 27.3 k 5.1). No patient had received any treatment since the onset of symp- toms. Fifty-four control subjects matched for age, sex and weight were also included in the study.
Design of the study The clinical diagnosis was established according to WHO criteria [2]. B-cell function was evalu- ated in subjects attending as outpatients. After an overnight fast patients were given an intravenous injection of 1 mg glucagon (Novo Industrie, Bagsvaerd, DK-2880 Denmark) and free-flowing venous blood was obtained before and 6 min after glucagon injection [9,10].
Laboratory determination Samples for plasma glucose determinations were collected in tubes containing sodium fluoride, centrifuged immediately and stored at - 20°C and analyzed using the glucose oxidase method on a glucose analyzer (Beckman Instruments, Anaheim, CA, U.S.A.). Serum, separated from freshly drawn venous blood, was stored at - 20’ C and used later for measurement in dupli-
cate tests of C-peptide immunoreactivity (CPR) and immunoreactive insulin (IRI) by radio- immunoassay (RIA) using Serono RIA Kits. Quality control of the analyses was checked using commercial quality control specimens run after every seven test samples. The mean values obtained.on the controls were within the values quoted by the manufacturer. Assay for CPR was carried out using an immunoprecipitation tech- nique with 20 y0 polyethyleneglycol solution and IRI was assayed by a double-antibody magnetic solid-phase suspension technique using an MAIA magnetic separator. The intra- and inter-assay coefficients of variation were 8.6% and 9.2% for C-peptide at a level of 0.58 k 0.05 ng/ml, and 11.7% and 10.6% for insulin at a level of 8.2 + 0.94 pU/ml.
BMI was computed as kilograms body weight divided by height squared (m’). BMI has been shown to correlate well with percent desirable weight determined from actuarial charts. Men with a BMI 227 and women with a BMI 225 were considered obese [2].
Data analysis All data were fed into a 60-megabyte AMDAHL computer and processing for the classification of
patients according to BMI was made using the SPS S-X statistical software package. Results are reported as mean k SD unless otherwise indi- cated.
Results
Age distribution and BMI The age range of the subjects studied is shown in Table 1; 56.5 % of patients were between 35 and 54 years. Only 13.4% were below the age of 35 years. One (0.5%) diabetic patient was found with an age of less than 15 years and 3 (1.4%) patients with ages over 75 years.
Based on BMI criteria obesity was more preva- lent in female (5 1%) 110 patients) than male sub- jects (12.5 %, 27 patients) (Table 1).
Response to glucagon test A summary of laboratory data for all patients grouped according to age is shown in Table 2. The highest frequency of patients falls between the age group 35-54 years. Plasma glucose concentra- tions increased in all groups from a mean + SD 14.9 & 1.7 mmol/l (range 11.9-17.1 mmol/l) to
TABLE 1
Characteristics of 216 newly diagnosed Saudi diabetic patients
53
15.2 + 1.4 mmol/l (range 14.0-17.2 mmol/l). Sta- tistical analysis on the increment of plasma glu- cose concentration after the glucagon stimulation test was performed. Mean k SD increments of plasma glucose were not significantly correlated with age distribution (r = 0.01, P -C 0.005). Fur- ther analysis was performed to study the correla- tion between glucose concentrations before and after a glucagon stimulation test. Fig. 1 shows fasting plasma glucose concentration plotted against maximum post-glucagon plasma glucose concentration. The correlation coefficient was strongly positive (r = 0.93, P < 0.001).
An indication of B-cell endogenous insulin secretion was obtained by measuring serum insulin and C-peptide, respectively, before and 6 min after the glucagon stimulation test. Insulin and C-peptide increased significantly in 187 (87%) of the patients studied (P < 0.01) from fasting values of 14.0 + 1.8 pU/ml and 1.8 + 0.4 ng/ml to peak values of 2 1.1 k 3 pU/ml and 2.4 2 0.4 ng/ml for insulin and C-peptide (Table 2). However, serum insulin and C-peptide levels were not affected by the age distribution (r = 0.01, P < 0.005). In the diabetic subjects we observed a wide spectrum of response with incre-
Age
Age range
(years)
Number of % patients patients n = 216 studied
BMI
BMI value
(kg/m* )
Number of patients n = 216
% patients studied
<15 1 0.5 15-24 4 1.8 25-34 24 11.1 35-44 62 28.1 45-54 60 27.8 55-64 35 16.2 65-74 27 12.5
215 3 1.4
227 <27
225 <25
Male 27 52
Female
110 27
12.5 24
51.0 12.5
Range ( 13-95) (16.2-51.3) Mean f SD 46.9 + 12.9 21.3 k 5.1
TA
BL
E
2A
Lab
orat
ory
data
of
the
21
6 ne
wly
di
agno
sed
Saud
i di
abet
ic
patie
nts
in r
elat
ion
to
age
Age
(y
ears
)
Glu
cose
(m
mol
/l)
Insu
lin
(pU
/ml)
Incr
emen
t Fa
stin
g
C-p
eptid
e (n
g/m
l)
Fast
ing
Peak
Pe
ak
Incr
emen
t Fa
stin
g Pe
ak
<15
15
-24
25 -
34
35-4
4
45 -
54
55 -
64
65 -
74
215
Mea
n f
SD
14.9
f 1.
7 15
.2+
1.
4 R
ange
(1
1.9
- 17
.1)
(14.
0 -
17.2
)
17.1
11
.9k5
.8
(5.2
-
16.1
) 14
.3 +
3.9
(8
.3 -
18
.5)
15.2
f 3.
9 (6
.8 -
22.
3)
15.9
f 4.
4 (6
.2 -
24.
3)
15.5
f 4.
3 (6
.1 -
22.
3)
16.1
+ 6
.0
(5.7
- 2
8)
13.1
f 2.
2 (1
1.2
- 15
.6)
17.2
12
.7 f
6.0
(5.7
-
16.2
) 14
.4 f
3.0
(9.7
-
20.9
) 15
.3 f
3.8
(7.8
- 2
5.4)
16
.0 f
4.2
(7.1
- 2
4.8)
15
.9 +
4.4
(6
.3 -
24
.2)
16.3
f 6.
0 (6
.1 -
28.
2)
14.0
f 2.
5 (1
1.5
- 16
.4)
0.1
0.8
f 0.
9
0.7
f 0.
6
0.8
+ 1
.1
0.5
+ 0
.5
0.7
f 0.
8
0.4
& 0
.3
0.8
+ 0
.5
11.0
15
.7 f
2.4
(12.
4 -
18.3
) 14
.6 f
5.5
(6 -
26.
8)
14.7
+ 7
.7
(2.2
-
39.8
) 13
.6 f
6.6
(3.6
- 3
2.4)
14
.2 f
4.9
(6.3
- 2
4.2)
16
.3 &
8.2
(5
.6 -
32
.2)
11.9
k8.4
(2
.6 -
18
.7)
24.7
24
.9 &
4.9
(1
9.5
- 29
.7)
22.3
f 9
.9
(6.7
-
36.6
) 22
.1 &
12.
8 (6
.4 -
55
.4)
17.6
f 8.
9 (1
.4-4
4.8)
21
.1 f
11
.9
(8.9
-
69.6
) 20
.1 f
11
.3
(1.1
-43
.6)
16.8
+ 1
0.1
(1.8
-2.2
)
0.6
f 0.
2 14
.0 +
1.8
21
.1+
3.
0 (1
1 -
16.3
) (1
1.8
- 24
.9)
Incr
emen
t
13.7
9.
2 +
6.6
7.9
f 6.
9
8.1
f. 9.
6
5.7
k 6.
2
7 +
10.
4
6.2
f 5.
6
4.9
f 1.
3
1.1
1.9
+ 0
.6
(1.1
- 2
.5)
2.1
k 0.
8 (1
.2-3
.6)
2.0
rt: 1
.0
(0.8
-
5.2)
1.
8 +
0.7
(0
.6 -
3.
8)
1.8
f 0.
8 (0
.9 -
3.
3)
2.3
f 1.
1 (0
.6 -
4.2
) 1.
6kO
.6
(1.2
- 2
.2)
2.4
3.0
* 3.
0 (2
.1 -
3.
9)
2.6
f 0.
9 (1
.2-3
.9)
2.5
* 1.
1 (1
.1 -
5.
5)
2.2
+ 0
.9
(0.8
- 4
.9)
2.4*
1.
1 (0
.9 -
4.9
) 2.
1 f
1.3
(1.0
-
5.2)
1.
8 &
0.6
(1
.2-2
.5)
7.5
f 3.
2 1.
8 f
0.4
2.4
& 0
.4
(1.1
- 2
.3)
(1.8
-
3.0)
1.3
1.1
* 0.
7
0.6
f. 0.
6
0.6
f 0.
5
0.4
f 0.
5
0.6
f 0.
7
0.5
f 0.
3
0.2
f 0.
2
0.7
+ 0
.4
The
mea
n f
SD a
nd t
he
rang
e of
fas
ting
plas
ma
gluc
ose,
in
sulin
an
d C
-pep
tide
conc
entr
atio
ns
for
the
patie
nts
stud
ied.
O
ne
hund
red
and
eigh
ty-s
even
(8
7%)
of
the
diab
etic
pa
tient
s re
spon
ded
to
a 1
mg
i.v.
gluc
agon
te
st.
TA
BL
E
2B
Lab
orat
ory
data
of
54
norm
al
cont
rol
subj
ects
in
rel
atio
n to
ag
e
Age
(y
ears
)
t15
15-2
4
25-3
4
35-4
4
45 -
54
55 -
64
65 -
74
215
Glu
cose
(m
mol
/l)
Insu
lin
(~U
jl)
C-p
eptid
e (n
g/m
l)
Fast
ing
Peak
In
crem
ent
Fast
ing
Peak
In
crem
ent
Fast
ing
Peak
In
crem
ent
- -
- _
_ -
- -
_ 5.
2 f
0.4
- -
- -
_ -
8.7
f 0.
7 2.
3kl.O
(4.9
-
5.8)
5.
8 +
1.2
1.
0 +
0.8
(7
.2 -
9.3
) 36
.1 f
7.
5 30
.4 +
18.
3 (2
.1 -
3.
0)
3.9
f 0.
9 1.
6 f
1.0
5.3
+ 0
.4
(5.1
- 6
.2)
14.9
+ 5
.6
(27.
7 -
55.5
) 2.
0 f
0.7
(3.0
-
5.4)
(4
.8 -
5.
8)
6.0
+ 1
.6
0.8
f 1.
1 (8
.5 -
25)
30
.1 f
8.
0 28
.2 k
10.
1 (1
.1 -
3.
0)
4.3
f 1.
0 2.
2 +
1.0
5.
5 f
0.6
(4.8
-
6.5)
11
.7k5
.6
(10.
1 -
36.6
) 2.
7 f
1.2
(1.2
-5.5
) (4
.7 -
6.6
) 6.
3 f
1.5
1.1
+ 0
.7
(6.1
- 2
2.5)
34
.5 f
7.
0 30
.2 f
7.0
(0.8
- 4
.6)
3.5
f 0.
8 2.
1 +
0.9
5.
6 k
1.1
(4.6
-
6.8)
12
.4 f
5.4
(12.
3 -
38.0
) 1.
7kO
.9
(1.2
-4.1
) (4
.4 -
7.
7)
(8.5
- 2
3.3)
(0
.7 -
3.
7)
5.6
f 0.
7 10
.2 +
3.1
2.
4 +
0.4
(5
.0 -
6.
8)
6.5
+ 1
.2
1.3
f 0.
8 (8
.0 -
16
.0)
29.9
f
16.4
16
.9 f
17.6
(1
.9 -
2.9
) 2.
8 +
0.4
1.
3kO
.7
5.8
f 1.
2 (4
.6 -
7.
2)
11.3
k2.8
(1
9.4
- 84
.8)
1.5
+ 0
.4
(2.0
-4.1
) (4
.1 -
7.
0)
(8.5
-
15.4
) -
- (1
.2-2
.0)
- -
_ _
_
Mea
n +
SD
5.
5 f
0.2
6.2
+ 0
.3
1.1
+ 0
.2
11.5
k2.1
32
.7 f
3.
1 26
.4 f
6.
4 2.
1 +
0.4
3.
6 +
0.6
1.
8 f
0.4
Ran
ge
(5.2
-
5.8)
(5
.8 -
7.
2)
(8.7
-
14.9
) (2
9.9
- 36
.1)
(1.5
-
2.7)
(2
.8 -
4.3
) ~
.~ _
___~
.
The
mea
n +
SD
and
the
ran
ge
of f
astin
g pl
asm
a gl
ucos
e,
insu
lin
and
C-p
eptid
e co
ncen
trat
ions
fo
r no
rmal
su
bjec
ts.
Ele
ven
subj
ects
re
spon
ded
to a
1 m
g i.v
. glu
cago
n
test
.
C-
PE
PT
IDE
b
efo
re
IV
GL
UtA
GO
N
(n9
/ml
)
GL
UC
OS
E
afte
r I V
. G
LU
CA
GO
N
(m
mo
l I
L )
C-P
EP
TID
E
afte
r IV
G
CU
CA
GO
N
(ng
lml
)
57
. Before IV glucogon
D After IV glucogon
< P -_ N-110 N= 27 N=27 N=52
\ P 6
8 4 .-
‘O :: 2
CJ 0
$ 60
3 40 3
5 20 a = 0
Fig. 5
BMI BMI BMI BMI
225 <25 227 <27
FEMALE MALE
The relationship between obesity (in terms of BMI values) and glucagon stimulatory effect on plasma glucose concentration and B-cell secretion. Patients were grouped based on their BMI. Filled circles (0) represent the highest and lowest pre-stimulation insulin, C-peptide and glucose levels and open circles (a) represent the highest and lowest
post-stimulation levels for each group.
ments from 4.9 to 13.7 PI-J/ml and 0.2 to 1.3 ng/ml in insulin and C-peptide values after stimulation, respectively. In response to glucagon injection (1 mg i.v.), a direct correlation was observed for insulin and C-peptide concentra- tions before and after the stimulation, with r = 0.59, P < 0.01 and r = 0.8, P < 0.01, respec- tively (Figs. 2 and 3). There was a negative corre- lation between peak serum insulin and C-peptide levels against fasting and post-glucagon glucose concentrations with r = - 0.18, P < 0.01 and r = - 0.16, P -C 0.02, respectively. Insulin and C-peptide concentrations had a positive correla- tion value before (Fig. 4) and after the glucagon test at r = 0.5, PC 0.01 and I = 0.44, P< 0.01, respectively.
Eleven of the 54 normal control subjects took the glucagon test. The 6-min post-glucagon value for glucose was 6.2 k 0.3 mmol/l (range
5.8-7.5 mmol/l), for insulin 32.7 k 3.1 pU/ml (range 29.9-36.1 pU/ml) and for C-peptide 3.6 + 0.6 ng/ml (range 2.8-4.3 ng/ml).
BMI and response to glucagon Although this study was not designed to deter- mine the relationship between obesity and the stimulatory effect of glucagon on B-cell secretion, it can be seen in Fig. 5 that obese patients, as defined by BMI [2], had higher maximum post- glucagon values (for glucose, insulin and C-peptide). In addition, a direct correlation was found between body size and B-cell secretory capacity (r = 0.67, P < 0.001). The mean k SD fasting insulin and C-peptide levels in the obese patients were 18.5 & 9.1 pU/ml and 2.44 k 0.8 ng/ml, respectively, while the non-obese group had lower levels, 13.0 f 5.0 pU/ml for insulin and 1.85 + 0.72 ng/ml for C-peptide. However, slight- ly less B-cell secretory capacity was observed in female than in male diabetic subjects. This may be due to the sample size distribution obtained.
Discussion
Of the newly diagnosed Saudi diabetic patients studied 63.5 y0 were women and 36.5 y0 were men, suggesting perhaps a higher prevalence of diabetes mellitus in women in Saudi Arabia. The peak onset of diabetes mellitus occurred above the age of 30 years in adults, in agreement with previous reports from the area [ 13,141. Among our patients 63% were overweight, as defined by BMI, com- pared with 22.2% of non-diabetic control sub- jects.
This present study shows our assessment of the B-cell secretory capacity in the newly diagnosed Saudi diabetic patients. Glucagon was used as a provocative test for B-cell reserve [ 151. Unger et al. [ 161 and Faber and Binder [9,17] have shown that diabetic patients as a whole have ele- vated C-peptide values (compared with normal control subjects) but which are consistent with hyperglycemia, and their peak C-peptide response
58
occurs within 4-15 min (maximum 6 min) after intravenous injection of glucagon.
The glucagon test, apart from being rapid and easy to perform, has a high predictive value for B-cell responsiveness. Eighty-seven percent (187/216) ofthe patients had a significant quantity of insulin (21.1 k 3 pU/ml) and C-peptide (2.4 f 0.4 ng/ml) levels in the plasma, indicating that these diabetic patients secrete C-peptide (and presumably insulin) under normal conditions. These findings are supported by a direct correla- tion between fasting and stimulated serum insulin and C-peptide concentrations in response to glucagon (Figs. 2 and 3). The explanation for the close correlations may be that, in individual patients, fasting and glucagon-stimulated C-pep- tide values are similarly dependent on the con- comitant blood glucose concentration. Our study is in agreement with previous investigations [ 9,12,18] with the exception of that of Mire1 et al. [ 191. In relation to plasma glucose levels our find- ings support a previous study [ 1 l] stating that no correlation exists between serum C-peptide and plasma glucose.
In the present series, we have found a very wide spectrum of post-glucagon increments for insulin (4.9-13.7 pU/ml) and for C-peptide (0.2-1.3 ng/ml). These increments in total insulin and C- peptide are almost certainly due to insulin and C-peptide alone as pro-insulin concentration is too small to interfere or to be detectable in the assay [20]. A similar study was conducted by Mire1 et al. [ 191 and the same range of concentra- tions was obtained. It was then claimed that allowances must be made for differences in assay (antibodies and separation techniques), body weight, diet, activity and the current lack of an international insulin or C-peptide standard. Our results are in agreement with this.
The obese subset of our patients (BMI 2 27 for males and BMI 2 25 for females) showed higher levels of fasting and post-glucagon stimulation plasma glucose, insulin and C-peptide levels than the normal-weight subjects. Furthermore, during the evaluation of the B-cell secretory reserve capacity the post-glucagon concentrations of
insulin and C-peptide were affected by body size (I = 0.67, P < 0.001) (Fig. 5). The slight dif- ferences in insulin and C-peptide values in male and female obese diabetic patients may be due to the sample size available when the study was conducted, so caution should be used in extra- polating these findings. This could be attributed to the fact that caloric intake was adjusted to corre- spond to body size [ 2 11. The reproducibility of the glucagon stimulation test was examined in a previous study [22], the outcome of which showed that the prevailing fasting blood glucose affected the results of the glucagon stimulation test done on type 1 (insulin-dependent) and type 2 (non-insulin-dependent) patients having under- gone treatment for 6 months to 11 years. Thus, our present findings should be confirmed after a period of treatment.
In conclusion, using a glucagon stimulation test, we have found a significant endogenous B-cell reserve in newly diagnosed Saudi diabetic patients. This confirms our supposition that the majority of adult Saudi diabetics are type 2 with obesity and insulin sensitivity as major associated factors. Direct evidence for differential insulin sensitivity should be examined in this population and in particular the presence of tissue sensitivity to insulin [23].
Acknowledgements
We thank Prof. K.G.M.M. Alberti for his con- tinuous valuable comments and Mr. Saleh Abdul- Mutaal of the Statistics department for his advice. We are indebted to Mr. N. Vigo and other techni- cians who participated in the whole assessment and to Mr. Z.A. Javed for typing the manuscript. This work was supported by Grant Bio/1404/38 from the Research Center, College of Science, King Saud University, Riyadh, Saudi Arabia.
References
1 National Diabetes Data Group (1979) Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 28, 1039-1057.
59
2 WHO Study Group on Diabetes Mellitus (1985) WHO Tech. Rep. Ser. No. 727.
3 O’Rahilly, S., Spivey, RX, Holman, R.R., Nugent, Z., Clark, A. and Turner, R.C. (1987) Type II diabetes of early onset: a distinct clinical and genetic syndrome. Br. Med. J. 294, 923-928.
4 Viswanathan, M., Mohan, V., Snehalatha, C. and Ramachandran, A. (1985) High prevalence of Type 2 (non-insulin-dependent) diabetes among the offspring of conjugal Type 2 diabetic parents in India. Diabetologia 28, 901-910.
5 Tattersall, R.B. (1974) Mild familial diabetes with domi- nant inheritance. Q.J. Med. 43, 339-357.
6 Fonseca, V., Laajam, M.A. and Tongia, R.K. (1985) Non- ketotic diabetes mellitus in young non-obese Saudi Arabians. East Afr. Med. J. 62, 21-25.
7 Rubenstein, A.H., Clark, J.L., Melanin, F. and Steiner, D.F. (1969) Secretion of proinsulin C-peptide by pan- creatic beta-cells and its circulation in blood. Nature 224, 697-699.
8 Heding, L.G., Larsen, U.D., Markussen, J., Jorgensen, K.H. and Hallund, 0. (1974) Radioimmunoassay for human pork and ox C-peptide and related substances. Horm. Metab. Res. 5 (Suppl.), 40-44.
9 Faber, O.K. and Binder, C. (1977) C-peptide response to glucagon: a test for the residual B-cell function in diabetes mellitus. Diabetes 26, 605-610.
10 Gjessing, H.J., Matzen, L.E., Froland, A. and Faber, O.K. (1987) Correlations between fasting plasma C-pep- tide, glucagon-stimulated plasma C-peptide, and urinary C-peptide in insulin-treated diabetics. Diabetes Care 10, 487-498.
11 Madsbad, S., Krarup, T., McNair, P., Christiansen, C., Faber, O.K., Transbol, I. and Binder, C. (1981) Practical clinical value of the C-peptide response to glucagon stimulation in the choice oftreatment in diabetes mellitus. Acta Med. Stand. 210, 153-156.
12 Welborn, T.A., Garcia-Webb, P. and Bonser, A.M. (198 1) Basal C-peptide in the discrimination of type I from type II diabetes. Diabetes Care 4, 616-619.
13 Bacchus, R.A., Bell, J.L., Madkour, M. and Kilshaw, B. (1982) The prevalence of diabetes melhtus in male Saudi Arabs. Diabetologia 23, 330-332.
14 Fatani, H.H., Mira, S.A. and El-Zubier, A.C. (1987) Prevalence of diabetes mellitus in rural Saudi Arabia. Diabetes Care 10, 180-183.
15 Samols, F., Marri, G. and Marks, V. (1965) Promotion of insulin secretion by glucagon. Lancet ii, 115-l 16.
16 Unger, R.H., Aydin, I., Nakabayashi, H., Srikanta, C.B. and Raskin, P. (1976) The effects of glucagon adminis- tration to non-diabetics and diabetics. Metabolism 25 (Suppl. I), 1523-1526.
17 Binder, C. and Faber, O.K. (1978) Residual beta-cell function and its metabolic consequences. Diabetes 27 (Suppl. l), 226-229.
18 Garcia-Webb, P., Bonser, A. and Welborn, T.A. (1982) Correlation between fasting serum C-peptide and B-cell insulin secretory capacity in diabetes mellitus. Diabetologia 22, 296.
19 Mirel, R.D., Ginsberg-Fellner, F., Horwitz, D.L. and Rayfield, E.J. (1980) C-peptide reserve in insulin-depen- dent diabetes: comparative responses to glucose, gluca- gon and tolbutamide. Diabetologia 19, 183-188.
20 Horwitz, D.L., Stan, J.I., Maki, M.E., Blackard, W.G. and Rubenstein, A.H. (1975) Proinsulin, insulin and C-peptide concentrations in human portal and peripheral blood. J. Clin. Invest. 55, 1278-1283.
21 Hoogwerf, B.J. and Goetz, F.C. (1983) Urinary C-pep- tide: a simple measure of integrated insulin production with emphasis on the effects of body size, diet, and corticosteroids. J. Clin. Endocrinol. Metab. 56, 60-67.
22 Arnold-Larsen, S., Madsbad, S. and Kuhl, C. (1987) Reproducibility of the glucagon test. Diabet. Med. 4, 299-303.
23 Bergman, R.N., Finegood. O.T. and Aden, M. (1985) Assessment of insulin sensitivity in vivo. Endocrinol. Rev. 6, 45-86.