Baseline clinical characteristics
There were no significant differences between the five groups in the baseline characteristics (Age, Male, Smoking, LVEF, Hypertension, Contrast volume, TG, TC, HDL-C, LDL-C, Hydration amount, b-blockers, ACEI/ARB, Diuretics, CCBs) before operation. Patients with elevated pre-procedural glucose levels had higher BMI (p = 0.005). In addition, patients with higher levels of glucose were more often on therapy with PCI. (p = 0.049) (Table 1)
Table 1
Comparisons of baseline characteristics between the five groups.
Variables
|
GLU
< 126mg/dl
(n = 29)
|
GLU
126 to150mg/dl
(n = 84)
|
GLU
151 to175mg/dl
(n = 163)
|
GLU
176 to 200mg/dl (n = 179)
|
GLU
> 200mg/dl
(n = 77)
|
P
|
Age (years)
Male (%)
Smoking (%)
LVEF
Hypertension (%)
BMI (kg/m2)
Contrast volume (mL)
TG (mmol/L)
TC (mmol/L)
HDL-C (mmol/L)
LDL-C (mmol/L)
Hydration amount (mL)
β-antagonist (%)
ACEI/ARB (%)
CCB (%)
Hypoglycemic agents (%)
Insulin (%)
PCI (%)
Basal Scr (mmol/L)
48h Scr(mmol/L)
72h Scr (mmol/L)
|
65.86 ± 6.65
16(55.2)
20(67.0)
60.59 ± 8.17
15(51.7)
23.76 ± 2.04
168.6 ± 41.55
1.84 ± 1.04
4.62 ± 0.41
1.29 ± 0.47
2.50 ± 0.69
1318.6 ± 266.92
21(72.4)
15(51.7)
3(7.8)
24(82.8)
11(37.9)
10(34.5)
98.92 ± 14.53
110.6 ± 11.92
107.3 ± 12.84
|
66.89 ± 6.94
40(47.6)
53(63.1)
60.42 ± 6.96
49(92.2)
23.98 ± 2.08
165.2 ± 49.44
1.69 ± 0.97
4.52 ± 0.38
1.26 ± 0.33
2.54 ± 0.61
1340.6 ± 267.91
70(83.3)
43(51.2)
10(11.9)
62(73.8)
32(38.1)
36(42.9)
102.07 ± 14.61
114.7 ± 18.48
110.1 ± 17.36
|
67.62 ± 6.27
91(55.8)
116(71.2)
59.56 ± 7.07
78(47.9)
24.63 ± 2.12
174.5 ± 59.79
1.76 ± 0.91
4.53 ± 0.43
1.24 ± 0.33
2.49 ± 0.58
1315.2 ± 236.53
131(80.4)
71(43.6)
19(11.7)
134(82.2)
61(72.6)
78(47.9)
101.78 ± 15.01
114.6 ± 17.69
109.0 ± 17.23
|
67.16 ± 7.43
86(48.0)
107(59.8)
60.73 ± 7.02
101(56.4)
24.78 ± 2.00
177.0 ± 59.06
1.90 ± 1.15
4.58 ± 0.56
1.25 ± 0.25
2.48 ± 0.52
1274.4 ± 278.33
160(89.4)
95(53.1)
25(14.0)
155(86.6)
79(44.1)
96(53.6)
102.8 ± 11.44
118.1 ± 16.07
115.3 ± 15.09
|
68.58 ± 7.24
40(51.9)
50(64.9)
59.26 ± 7.18
43(55.8)
24.84 ± 2.06
182.7 ± 65.43
1.76 ± 0.88
4.50 ± 0.48
1.33 ± 0.55
2.57 ± 0.54
1307.1 ± 255.91
65(84.4)
40(51.9)
10(13.0)
63(81.8)
30(40.0)
47(61.0)
108.1 ± 14.08
126.8 ± 13.86
118.6 ± 14.42
|
0.345
0.605
0.268
0.611
0.452
0.005*
0.366
0.566
0.538
0.425
0.793
0.259
0.077
0.472
0.963
0.168
0.750
0.049*
0.005*
0.000*
0.000*
|
Table 1. Data are expressed as mean ± SD or n (%). ACEI: angiotensin-converting enzyme inhibitor; ARB: angiotensin receptor blocker; BMI: body mass index; TC: total cholesterol; TG: triglyceride; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; LVEF: left ventricular ejection fraction; CCB: calcium channel blockers; PCI: Percutaneous Coronary Intervention.
Comparison of Scr level and CIN incidence in each group
Patients with elevated pre-procedural glucose had higher basal Scr (p = 0.005), higher Scr values at 48 h and 72 h after CAG/PCI (p < 0.0001, Table 1). Multivariable logistic regression analysis was used to analyze the factors affecting CIN. CIN was taken as the dependent variable, and factors that may affect the development of CIN (male, age, Contrast volume, Hydration amount, BMI, Hypertension, PCI) were taken as independent variables. CIN was used as the dependent variable to exclude confounding factors. (OR = 1.008(1.001–1.015), p = 0.033)
The multivariate analysis confirmed the association between pre-procedural glucose and the risk of CIN after adjustment for baseline confounding factors. The results showed that patients with elevated pre-procedural glucose (above the median value 158mg/dl) in the main high-risk subgroup had a higher risk of CIN, such as BMI (BMI > 23.9kg/m2: adjusted OR = 1.460(0.722–2.953), p = 0.292; BMI ≤ 23.9kg/m2: adjusted OR = 5.517(1.576–19.315), p = 0.008), Age (Age ≥ 65: adjusted OR = 2.947(1.285–6.761), p = 0.011; Age < 65: adjusted OR = 1.871(0.762–4.596), p = 0.172), PCI (PCI: adjusted OR = 2.983(1.212–7.343), p = 0.017; No PCI: adjusted OR = 1.654(0.714–3.843), p = 0.240).(Fig. 2)
The trend test was used to further demonstrate the relationship between pre-procedural glucose levels and CIN incidence. It can be seen from the changes in Hazard Ratio (HR) (1.000,1.646,2.285,3.227,4.788), with the increase of pre-procedural glucose levels, the incidence of CIN increases gradually with a significant trend (p < 0.0001). (Table 2)
Table 2
The logistic regression analysis and the trend test
Blood glucose
|
Participants, n
|
CIN, n
|
Adjusted Hazard Ratio
|
Model 1
|
Model 2
|
Model 3
|
GLU < 126mg/dl
GLU126 to150mg/dl
GLU151 to175mg/dl
GLU176 to 200mg/dl
GLU > 200mg/dl
P for trend
|
29
84
163
179
77
--
|
2
9
23
36
20
--
|
1.000(Reference)
1.620(0.329–7.976)
2.331(0.520-10.449)
3.399(0.772–14.961)
4.737(1.032–21.744)
< 0.0001*
|
1.000(Reference)
1.660(0.332–8.301)
2.368(0.520-10.783)
3.250(0.726–14.549)
4.710(1.004–22.096)
< 0.0001*
|
1.000(Reference)
1.646(0.331–8.180)
2.285(0.503–10.377)
3.227(0.724–14.394)
4,788(1.026–22.351)
< 0.0001*
|
Table 2. Model 1 was not adjusted. Model 2 was adjusted for male, hypertension, LVEF, hydration amount and PCI. Model 3 additionally was adjusted for BMI.