Basic HTML Version
VOLUME 9 NUMBER 1 • MARCH 2012
23
SA JOURNAL OF DIABETES & VASCULAR DISEASE
RESEARCH ARTICLE
Table 1.
Clinical, anthropometric and biochemical characteristics of control
and diabetic subjects.
Parameters
Controls (
n
= 56) Type 2 diabetes (
n
= 37)
Age (years)
38 ± 6
42 ± 6
Body mass index (kg/m²)
25.4 ± 3.9
24.3 ± 3.0
Waist-to-hip ratio
0.90 ± 0.05
0.94 ± 0.04* (p = 0.004)
Neck circumference (cm)
35.4 ± 3.0
36.2 ± 2.8
MUAC (mm)
296 ± 28
275 ± 57
Triceps (mm)
14.4 ± 5.2
11.4 ± 3.4* (p = 0.03)
Body fat mass (%)
29.2 ± 6.2
27.6 ± 25.8
Systolic blood pressure
(mmHg)
114 ± 8
121 ±16
Diastolic blood pressure
(mmHg)
76 ± 8
83 ± 8* (p = 0.008)
Fasting glucose (mmol/l)
5.1 ± 0.4
8.0 ± 2.0* (p = 0.004)
2-hour glucose (mmol/l)
5.8 ± 1.1
15.9 ± 3.5* (p = 0.004)
Triglycerides (mg/dl)
149 (52–376)
222 (54–532)* (p = 0.008)
Total cholesterol (mg/dl)
188 (90–261)
210 (161–278)* (p = 0.01)
HDL cholesterol (mg/dl)
31 (18–59)
30 (17-42)
LDL cholesterol (mg/dl)
127 (45–194)
135 (85–175)
Fasting insulin (pmol/l)
50 (7–155)
67 (25–145)
HOMA B
99 (21–187)
50 (13–141)* (p = 0.004)
HOMA S
88 (29–554)
61 (30–160)* (p = 0.04)
TNF-
α
(pg/ml)
9.9 (4.8–28.9)
12.3 (5.8–45.1)* (p = 0.03)
IL-6 (pg/ml)
2.8 (1.1–10.6)
4.0 (1.9–22.5)* (p = 0.03)
*
p
< 0.05, significantly different compared to controls when using Student’s
t
-test.
MUAC = mid-upper arm circumference; HOMA B = B-cell function and
HOMA S = insulin sensitivity, assessed by homeostasis model assessment;
TNF-
α
= tumour necrosis factor-alpha; IL-6 = interleukin-6.
Table 2.
Binary logistic regression analysis of TNF-
α
and IL-6 adjusted for
confounding factors (using control as reference).
Group
Factors
B
SE p-value Exp (B)
Type 2 diabetes
TNF-
α
0.103 0.047 0.026
1.109
BMI
0.048 0.065 0.461
1.049
Age
0.137 0.051 0.007
1.147
Gender
–0.063 0.756 0.933
0.939
Constant
–8.539 3.442 0.013
0.000
Type 2 diabetes
IL-6
0.155 0.092 0.094
1.167
BMI
–0.178 0.102 0.080
0.837
Age
0.182 0.056 0.001
1.200
Gender
–0.293 0.707 0.679
0.746
Constant
–3.406 2.997 0.256
0.033
Methods
This retrospective analytical case–control study was conducted
in the Research Division, Bangladesh Institute of Research and
Rehabilitation in Diabetes, Endocrine and Metabolic Disorders
(BIRDEM), Dhaka. A group of 37 subjects with type 2 diabetes was
selected from the outpatient department of BIRDEM. A group of 56
age-, gender- and body mass index (BMI)-matched healthy subjects
without a family history of diabetes was selected as controls from
among the circle of friends of the diabetic subjects, considering
them to be of the same socio-economic status. Written consent
was obtained from all volunteers.
A clinical history was taken by a registered physician using a
self-administered, pre-designed questionnaire. Anthropometric
measurements were carried out using standard methods. The
measurement of height and weight was done with light clothes
and without shoes. The weighing scale was calibrated daily with
a known standard weight. Triceps, waist, hip and mid-upper
arm circumference circumferences were measured with the tape
measure provided by the nutritionist. Body fat mass (%) was
measured using a body fat monitor, HBF 302, Omron Corp, Japan.
Blood pressure was taken with a standard blood pressure machine
after a 10-min rest.
After overnight fasting, subjects were requested to come on
a pre-scheduled morning for the fasting blood sample. Subjects
were then given 75 g anhydrous glucose dissolved in 250 ml water.
Blood was taken by venepuncture in the fasting condition and two
hours after glucose loading.
Serum glucose, total cholesterol, triglyceride, high-density
lipoprotein (HDL) and low-density lipoprotein (LDL) cholesterol
levels were determined by enzymatic colorimetric methods using
commercial kits (Randox Laboratories Ltd, UK). Serum insulin levels
were determined using the enzyme-linked immunosorbent assay
(ELISA) method (Linco Research Inc, USA). Serum TNF-
α
and IL-6
concentrations were measured by the solid-phase, enzyme labelled,
chemiluminescent immunometric assay (IMMULITE, DPC). Using
HOMA-CIGMA software, insulin secretory capacity (HOMA B) and
insulin sensitivity (HOMA S) were calculated from fasting glucose
and fasting insulin levels.
14
SPSS (Statistical Package for Social Science) software for
Windows version 10 (SPSS Inc, Chicago, Illinois, USA) was used
for statistical analysis. Data are expressed as mean ± SD (standard
deviation), median (range), and/or percentage (%), as appropriate.
A
p
-value was assessed by ANOVA or Mann-Whitney
U
-test (as
appropriate). Logistic and multiple regressions were done among
the parameters. A two-tailed
p
-value of < 0.05 was considered
statistically significant.
Results
Subjects with type 2 diabetes had significantly higher waist-
to-hip ratios (WHR) (
p
= 0.004), triceps skin-fold thickness (
p
=
0.03), diastolic blood pressure (
p
= 0.008) but not systolic blood
pressure, than the controls. Fasting serum triglyceride (TG) and
total cholesterol levels were significantly higher in the diabetic
subjects compared to controls (
p
= 0.008 and 0.01, respectively).
No significant differences in HDL and LDL cholesterol levels were
shown in the diabetic subjects compared to the controls. HOMA
B and HOMA S were significantly lower in the diabetic (
p
= 0.004,
p
= 0.04) subjects compared to controls. The diabetic subjects had
significantly higher levels of serum TNF-
α
(
p
= 0.032) and IL-6 (
p
=
0.032) than the controls (Table 1).
On binary logistic regression analysis (Table 2), it was found
that TNF-
α
was positively associated (
p
= 0.026) with the diabetic
subjects when the controls were used as the reference value, and
age, gender and BMI were adjusted for. Such an association of IL-6
with the diabetic state was not evident.