1. Parameters of patients:
Data from this study has found that, compared with T2DM group, the age of patients in T2DM+CHD group were older (60.12 ± 11.58 vs. 50.69 ± 13.00, P<0.05) and the course of T2DM were longer [4.00 (10.00,15.00) vs. 3.00 (0.52,10.00), P<0.05]. (Table 1).
Table 1. Comparison of general parameters between T2DM and T2DM+CHD group (mean ± SD) / [M(P25,P75)
Parameters
|
T2DM
|
T2DM+CHD
|
z/t/X2 value
|
P value
|
age(year)
|
50.69±13.01
|
60.12±11.58
|
3.814
|
0.001
|
BMI (kg/m2)
|
25.88±4.10
|
25.96±3.46
|
0.107
|
0.915
|
course of disease (year)
|
4.00(10.00,15.00)
|
3.00(0.52,10.00)
|
3.557
|
0.001
|
gender
|
male 26(54.2)
|
27(52.9)
|
0.015
|
0.903
|
female 23(45.8)
|
24(47.1)
|
smoking
|
no34(70.8)
|
38(74.5)
|
0.169
|
0.681
|
Yes14(29.2)
|
13(25.5)
|
alcohol drinking
|
no32(66.7)
|
37(72.5)
|
0.939
|
0.625
|
yes16(33.4)
|
14(27.4)
|
BMI: body mass index.
In the present study, compared with T2DM patients, the circulating level of homocysteine was significantly increased [17.60 (13.50,20.78) vs. 13.90 (10.25,15.83), P<0.05], while the level of vitamin D was significantly decreased (16.47 ± 3.82 vs. 19.87 ± 7.30, P<0.05) in T2DM +CHD patients. There were no significant differences found in plasma HbA1c, TC, TG, LDL-C and HDL-C levels between two groups. (Table 2).
Table 2. Comparison of blood parameters between T2DM and T2DM+CHD group (mean ± SD) / [M(P25,P75)]
Parameters
|
T2DM
|
T2DM+CHD
|
z/t/X2 value
|
P value
|
HbA1c (%)
|
8.87±2.44
|
8.99±2.33
|
0.250
|
0.800
|
CRP (mg/L)
|
1.59(1.11,2.63)
|
1.18(0.81,1.81)
|
-0.650
|
0.516
|
TC (mmol/L)
|
5.11±1.05
|
4.80±1.36
|
-1.540
|
0.124
|
TG (mmol/L)
|
1.70(1.20,2.55)
|
1.50(1.22,2.70)
|
-0.240
|
0.810
|
HDL-C (mmol/L)
|
1.25±0.29
|
1.18±0.30
|
-1.040
|
0.299
|
LDL-C (mmol/L)
|
3.03±0.67
|
2.95±0.93
|
-0.538
|
0.592
|
VLDL (mmol/L)
|
0.77(0.55,1.16)
|
0.70(0.55,1.23)
|
-0.612
|
0.541
|
Homocysteine (umol/L)
|
13.90(10.25,15.83)
|
17.60(13.50,20.78)
|
-3.788
|
0.001
|
Uric acid (umol/L)
|
347.64±86.26
|
337.63±116.94
|
-0.480
|
0.633
|
Blood creatinine (umol/L)
|
70.08±22.91
|
75.76±32.68
|
0.990
|
0.325
|
Folic acid (ng/ml)
|
9.30±4.73
|
10.69±5.26
|
1.080
|
0.282
|
Vitamin B12 (ng/ml)
|
368.75(261.05,591.03)
|
245.20(189.48,415.05)
|
-1.969
|
0.049
|
25-hydroxy vitamin D (ng/ml)
|
19.87±7.30
|
16.47±3.82
|
-2.850
|
0.006
|
HbA1c: hemoglobin A1c; CRP: C-reactive protein; TC: total cholesterol; TG: triglycerides ; HDL: high-density lipoprotein; LDL: low-density lipoprotein; VLDL: very low-density lipoprotein.
2. Plasma Angptl7 and Fet-A levels
In our study, the plasma Angptl7 level was significantly higher in T2DM+CHD patients than that in T2DM patients [1.55(0.94,2.57)ng/ml vs. 0.96(0.36,1.34)ng/ml,P<0.05] (Fig. 1). While the plasma Fet-A level was significantly lower in T2DM+CHD patients than that in T2DM patients (168.99 ± 25.80 ng/ml vs.193.85 ± 30.04 ng/ml, P < 0.05) (Fig. 2).
3. Correlation analysis of plasma Angptl7 and Fet-A with the clinical and biochemical characteristics of the study participants
Spearman correlation analysis showed that plasma Angptl7 levels were positively correlated with age, HbA1c, folic acid, and vitamin B12, while plasma Fet-A level was negatively correlated with vitamin B12 level (Table 3). No significant correlation was found between plasma Angptl7 and Fet-A levels (r=-0.194,P=0.054) (Fig. 3). Multiple stepwise regression analysis showed that, age, folic acid and vitamin B12 levels were influence factors for Angptl7 (P all < 0.05) while cholesterol and vitamin B12 were influence factors for Fet-A (P all < 0.05) (Table 4).
Table 3. Analysis of factors associated with plasma Angptl7 and Fet-A
variable
|
Angptl7
|
Fet-A
|
r/rs
|
P
|
r/rs
|
P
|
age(year)
|
0.331
|
0.001
|
-0.009
|
0.931
|
course of T2DM(year)
|
0.065
|
0.525
|
-0.165
|
0.103
|
BMI(kg/m2)
|
0.047
|
0.649
|
0.011
|
0.915
|
HbA1c(%)
|
0.214
|
0.037
|
-0.083
|
0.423
|
CRP (mg/L)
|
0.131
|
0.200
|
-0.083
|
0.423
|
TC (mmol/L)
|
0.090
|
0.379
|
-0.111
|
0.275
|
TG (mmol/L)
|
-0.010
|
0.991
|
0.040
|
0.696
|
HDL-C (mmol/L)
|
0.144
|
0.158
|
-0.101
|
0.323
|
LDL-C (mmol/L)
|
0.099
|
0.334
|
-0.172
|
0.091
|
VLDL (mmol/L)
|
-0.006
|
0.955
|
0.046
|
0.656
|
Homocysteine (umol/L)
|
0.063
|
0.545
|
0.084
|
0.416
|
Uric acid (umol/L)
|
-0.076
|
0.456
|
0.136
|
0.183
|
Blood creatinine (umol/L)
|
0.080
|
0.436
|
0.144
|
0.157
|
Folic acid (ng/ml)
|
0.499
|
0.000
|
-0.060
|
0.063
|
Vitamin B12 (ng/ml)
|
0.413
|
0.001
|
-0.295
|
0.016
|
25-hydroxy vitamin D (ng/ml)
|
-0.054
|
0.595
|
0.137
|
0.179
|
BMI: body mass index; HbA1c: hemoglobin A1c; CRP: C-reactive protein; TC: total cholesterol; TG: triglycerides; HDL: high-density lipoprotein; LDL: low-density lipoprotein; VLDL: very low-density lipoprotein.
In the present study, the correlation of plasma level of Angptl7 and Fet-A was measured by spearman correlation analysis. No significant correlation was found between plasma Angptl7 and Fet-A concentrations(r=-0.194, P=0.054).
Table 4. Multiple linear regression analysis of factors influencing plasma Angptl7 and Fet-A levels
variable
|
β
|
SE
|
β′
|
t
|
P
|
Angptl7
|
age(year)
|
0.156
|
0.053
|
0.498
|
2.930
|
0.006
|
Folic acid (ng/ml)
|
0.405
|
0.143
|
0.505
|
2.827
|
0.008
|
Vitamin B12 (ng/ml)
|
0.004
|
0.002
|
0.305
|
2.211
|
0.033
|
Fet-A
|
TC (mmol/L)
|
44.871
|
24.892
|
1.637
|
1.803
|
0.045
|
Vitamin B12 (ng/ml)
|
-0.022
|
0.014
|
-0.278
|
1.559
|
0.047
|
TG: triglycerides.
4. Logistic regression analysis and Receiver-operating characteristic curve (ROC) analyses of plasma Angptl7 and Fet-A
Logistic regression analysis showed that, age, plasma level of Angptl7 and Fet-A were the influence factors for occurrence of CHD in patients with T2DM (Table 5).
Data from ROC analysis has indicated that, for Angptl7, the area under the curve (AUC)was 0.667 (95% CI: 0.560~0.773, P<0.05) , and the cut point was 1.343ng/ml, the sensitivity and specificity was 58.8% and 72.9% separately. The AUC of Fet-A was 0.736 (95% CI: 0.638~0.883, P<0.05), and the cut point was 184.411 ng/ml, the sensitivity was 74.5% and the specificity was 62.5%. The AUC of Angptl7 and Fet-A combination was 0.750 (95% CI: 0.655~0.846, P<0.05), in which the sensitivity was 80.4% and the specificity was 60.4%. Therefore, the combination of two factors exhibited better predictive value for CHD in T2DM (Fig. 4).
Table 5. Logistic regression analysis of factors influencing the occurrence of CHD
variable
|
β
|
SE
|
Waldc2
|
P
|
OR
|
95%CI
|
Fet-A
|
0.042
|
0.017
|
6.209
|
0.013
|
1.043
|
1.009~1.078
|
age(year)
|
-0.152
|
0.052
|
8.479
|
0.004
|
0.859
|
0.775~0.951
|
variable
|
β
|
SE
|
Waldc2
|
P
|
OR
|
95%CI
|
Angptl7
|
-0.004
|
0.002
|
4.087
|
0.043
|
0.996
|
0.991~0.999
|
age(year)
|
-0.128
|
0.046
|
7.719
|
0.005
|
0.880
|
0.804~0.963
|