Table 1 Characteristics of the analyzed study population.

From: Early prediction of acute necrotizing pancreatitis by artificial intelligence: a prospective cohort-analysis of 2387 cases

Variable

Value (n = 2387)

Age in years, median (IQR)

57 (44–69)

Male, n (%)

1357 (56.85%)

BMI, median (IQR)

27.14 (23.88–31.25)

Etiology, n (%)

Biliary

955 (40.01%)

Alcoholic

484 (20.28%)

Hypertriglyceridaemia

81 (3.39%)

Biliary and alcoholic

39 (1.63%)

Biliary and hypertriglyceridaemia

13 (0.54%)

Alcoholic and hypertriglyceridaemia

58 (2.43%)

Post-ERCP

67 (2.81%)

Idiopathic

432 (18.10%)

Other

258 (10.81%)

Revised Atlanta classification

Mild, n (%)

1714 (71.81%)

Moderate, n (%)

551 (23.08%)

Severe, n (%)

122 (5.11%)

Mortality, n (%)

66 (2.76%)

Length of stay in days, median (IQR)

8 (6–12)

Patients with local complication, n (%)

623 (26.19%)

APFC, n (%)

510 (21.37%)

Pseudocyst, n (%)

179 (7.50%)

Acute necrotic collection, n (%)

233 (9.76%)

Patients with systemic complication, n (%)

202 (8.46%)

Respiratory failure, n (%)

136 (5.70%)

Heart failure, n (%)

52 (2.18%)

Renal failure, n (%)

83 (3.48%)

New-onset diabetes, n (%)

75 (3.14%)

  1. APFC acute peripancreatic fluid collection, BMI body mass index, ERCP endoscopic retrograde cholangiopancreatography, IQR interquartile range.