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 regarding the anti-diabetic therapy.
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 the 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.