Characteristics of Study Subjects
Based on the inclusion and exclusion criteria, 705 subjects, including 349 AP patients and 356 non-AP patients, were enrolled in our study. Table 1 shows the characteristics of the subjects. No significant differences between AP patients and non-AP patients were found with respect to age (p=0.989), gender (p=0.923), hepatitis C (p=0.349) and Apo B (p=0.198). However, we noted that AP patients were significantly associated with cigarette smoking (p<0.001), alcohol consumption (p<0.001), biliary stones history (p<0.001), hepatitis B (p=0.049), and BMI (p<0.001) and TC (p<0.001), TG (p<0.001), HDL (p<0.001), LDL (p=0.030), Apo A (p<0.001), and FPG (p<0.001) than controls. In patients with diabetes mellitus, 27 patients (79.4%) received anti-diabetic therapy in control group, while 59 patients (64.8%) received treatment in case group. There were no significant differences (p=0.118) between the two groups.
Table 1. Basic Characteristics
|
Non-AP patients
(n=356)
|
AP patients
(n=349)
|
p
|
Mean age, years (SD)
|
51.2±15.9
|
51.2±15.9
|
0.989
|
Gender, n (%)
|
|
|
|
Female
|
140 (39.3)
|
136 (39.0)
|
|
Male
|
216 (60.7)
|
213 (61.0)
|
0.923
|
Smoking history, n (%)
|
62 (17.4)
|
125 (35.8)
|
<0.001
|
Alcohol drinking history, n (%)
|
53 (14.9)
|
101 (28.9)
|
<0.001
|
Biliary stones, n (%)
|
8 (2.3)
|
101 (28.9)
|
<0.001
|
Hepatitis B , n (%)
|
17 (4.8)
|
7 (2.0)
|
0.049
|
Hepatitis C, n (%)
|
3 (0.8)
|
1 (0.3)
|
0.349
|
BMI, kg/m2 (IQR)
|
23.8±3.5
|
24.8±3.7
|
<0.001
|
TC, mg/dL (IQR)
|
178.0 (153.7, 209.3)
|
180.3 (145.6, 235.1)
|
<0.001
|
TG, mg/dL (IQR)
|
101.8 (77.0, 157.5)
|
131.9 (81.4, 349.6)
|
<0.001
|
HDL, mg/dL (IQR)
|
99.1 (86.7, 117.7)
|
90.3 (74.3, 112.4)
|
<0.001
|
LDL, mg/dL (IQR)
|
269.9 (224.8, 318.6)
|
258.4 (198.2, 236.3)
|
0.030
|
Apo A, g/L (IQR)
|
1.2 (1.1, 1.4)
|
1.1 (0.9, 1.3)
|
<0.001
|
Apo B, g/L (IQR)
|
0.9 (0.7, 1.0)
|
0.8 (0.7, 1.0)
|
0.198
|
FPG, mg/dL (IQR)
|
97.8 (89.4, 112.1)
|
132.8 (107.4, 172.7)
|
<0.001
|
Anti-diabetic therapy, n (%)
|
27 (79.4)
|
59 (64.8)
|
0.118
|
Data were numbers and percentages, or median (25th, 75th percentile), as appropriate.
Abbreviations: n, number; IQR, interquartile range; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); TC, total cholesterol; TG, triglyceride; HDL, high density lipoprotein; LDL, low density lipoprotein; Apo A, apolipoprotein A; Apo B, apolipoprotein B; FPG, fasting plasma glucose.
SI conversion factors: TC: 1 mg/dL =0.0259 mmol/L; HDL, LDL, TG:1 mg/dL= 0.0113 mmol/L; FPG:1mg/dL=0.0555mmol/L.
Effect of Metabolic Syndrome Components on AP
In univariate analysis, both of increased values of TG and decreased values of HDL were associated with occurrence of AP (OR=2.313; 95% CI 1.690-3.167, p<0.001; OR=0.582; 95% CI 0.431-0.786, p<0.001) (Table 2). Additionally, significant differences were observed between the two groups in patients with obesity, hyperglycaemia and hypertension(OR=1.608; 95% CI 1.186-2.181, p=0.002; OR=3.209; 95% CI 2.112-4.876, p<0.001; OR=1.473; 95% CI 1.089-1.991, p=0.012). In multivariate logistic regression models, after adjustment for smoking and alcohol drinking history, biliary stones history, Apo A and the components of metabolic syndrome, the results revealed that AP was associated with smoking history; alcohol consumption history; biliary stones history; elevated levels of TC, TG; hyperglycaemia and lower values of Apo A (OR=2.441; 95% CI 1.865-5.172, p<0.001; OR=1.777; 95% CI 1.060-2.977, p=0.029; OR=28.995; 95% CI 13.253-63.435, p<0.001; OR=1.992; 95% CI 1.246-3.183, p=0.004; OR=2.134; 95% CI 1.403-3.245, p<0.001; OR=2.261; 95% CI 1.367-3.742, p=0.001 and OR=0.270; 95% CI 0.163-0.447, p<0.001, respectively). However, obesity was not observed to be associated with the occurrence of AP (p=0.246) (shown in Table 3). After adjusting for smoking and alcohol drinking history, biliary stones, the prevalence of metabolic syndrome was more common in AP patients (30.9%) than in those without AP (13.2%) (OR=2.837; 95% CI 1.873-4.298, p<0.001) (shown in Table 4). As shown in Table 5 and Figure 1, for all AP patients, increased values of TG and low Apo A and FPG predicted AP with statistical significance (p<0.001, p<0.001, p<0.001); their AUCs were 0.620, 0.679 and 0.767, respectively. Among the three indicators, FPG had the best sensitivity (67.54%), and TG had the best specificity (90.17%) when the indicators were at their best cut-off values (shown in Table 5).
Table 2. Univariate Analysis Examining the Components of the Metabolic Syndrome
|
Non-AP patients
(n=356)
|
AP patients
(n=349)
|
OR
(95% CI)
|
p
|
TG≥150 mg/dL, n (%)
|
97 (27.2)
|
162 (46.4)
|
2.313 (1.690-3.167)
|
<0.001
|
HDL≥40 mg/dL (M) or ≥50 mg/dL (F), n (%)
|
183 (51.4)
|
133 (38.1)
|
0.582 (0.431-0.786)
|
<0.001
|
Obesity, n (%)
|
120 (33.7)
|
157 (45.0)
|
1.608 (1.186-2.181)
|
0.002
|
Hyperglycaemia, n (%)
|
37 (10.4)
|
93 (26.7)
|
3.209 (2.112-4.876)
|
<0.001
|
Hypertension, n (%)
|
129 (36.2)
|
159 (45.6)
|
1.473 (1.089-1.991)
|
0.012
|
Abbreviations: OR, odds ratio; TG, triglyceride; HDL, high density lipoprotein.
SI conversion factors: HDL, TG:1 mg/dL= 0.0113 mmol/L;
Table 3. Multivariate Analysis Examining the Components of the Metabolic Syndrome
|
Non-AP patients
(n=356)
|
AP patients
(n=349)
|
OR
(95% CI)
|
p
|
Smoking, n (%)
|
62 (17.4)
|
125 (35.8)
|
2.441 (1.865-5.172)
|
<0.001
|
Alcohol drinking, n (%)
|
53 (14.9)
|
101 (28.9)
|
1.777 (1.060-2.977)
|
0.029
|
Biliary stone, n (%)
|
8 (2.3)
|
101 (28.9)
|
28.995 (13.253-63.435)
|
<0.001
|
Hepatitis B, n (%)
|
17 (4.8)
|
7 (2.0)
|
0.439 (0.145-1.327)
|
0.145
|
Obesity, n (%)
|
120 (33.7)
|
157 (45.0)
|
1.258 (0.854-1.855)
|
0.246
|
TC≥220 mg/dL, n (%)
|
61 (17.1)
|
106 (30.4)
|
1.992 (1.246-3.183)
|
0.004
|
TG≥150 mg/dL, n (%)
|
97 (27.2)
|
162 (46.4)
|
2.134 (1.403-3.245)
|
<0.001
|
HDL≥40 mg/dL (M)
or
≥50 mg/dL (F) , n (%)
|
183 (51.4)
|
133 (38.1)
|
0.784 (0.522-1.176)
|
0.240
|
Apo A≥1 g/L, n (%)
|
321 (90.2)
|
230 (65.9)
|
0.270 (0.163-0.447)
|
<0.001
|
Hyperglycaemia, n (%)
|
37 (10.4)
|
93 (26.7)
|
2.261 (1.367-3.742)
|
0.001
|
Hypertension, n (%)
|
129 (36.2)
|
159 (45.6)
|
1.189 (0.817-1.732)
|
0.366
|
Abbreviations: OR, odds ratio; TC, total cholesterol; TG, triglyceride; HDL, high density lipoprotein; Apo A, apolipoprotein A.
SI conversion factors: TC: 1 mg/dL =0.0259 mmol/L ; HDL, TG:1 mg/dL= 0.0113 mmol/L;
Table 4. Association Between the Metabolic Syndrome and AP
|
Prevalence of Metabolic syndrome
|
OR
(95%CI)
|
p
|
AP patients, n (%)
|
108 (30.9)
|
2.837 (1.873-4.298)
|
<0.001
|
Non-AP patients, n (%)
|
47 (13.2)
|
|
|
Abbreviations: OR, odds ratio; AP, acute pancreatitis; CI, confidence interval.
Table 5. Identification of TG, Apo A and FPG in the patients with AP
Variable
|
AUC
|
p
|
Cut-off
|
Sensitivity(%)
|
Specificity(%)
|
+LR
|
-LR
|
TG, mg/dL
|
0.620
|
<0.001
|
218.6
|
39.08
|
90.17
|
3.98
|
0.68
|
Apo A, g/L
|
0.679
|
<0.001
|
1.15
|
63.16
|
66.85
|
1.91
|
0.55
|
FPG, mg/dL
|
0.767
|
<0.001
|
115.1
|
67.54
|
78.93
|
3.21
|
0.41
|
Abbreviations: AUC, area under curve; LR, likelihood ratio; TG, triglyceride; Apo A, apolipoprotein A; FPG, fasting plasma glucose.
SI conversion factors: TG:1 mg/dL= 0.0113 mmol/L; FPG:1mg/dL=0.0555mmol/L.
Association Between the Number of Metabolic Syndrome Components and AP
As shown in Figure 2, the incidence rate of AP obviously increased when there were more than three components of metabolic syndrome present. Moreover, the incidence rate of AP significantly declined when there were no components of metabolic syndrome present.