Шкала стратифікації ризику гострого панкреатиту після ендоскопічної ретроградної холангіопанкреатографії у пацієнтів з біліарною обструкцією

Abstract

Endoscopic retrograde cholangiopancreatography (ERCP) plays an im-portant role in management of biliary disorders. Post-ERCP pancreatitis (PEP) is the most common adverse event after ERCP.Purpose – to develop a practical reliable model for PEP prediction before ERCP in patient with signs of biliary obstruction.Materials and Methods. We analysed the data of 520 patients with common bile duct (CBD) ≥ 8 mm, who underwent ERCP. Patient-related factors were taken into the logistic regression model: gender, age (< 60 or ≥ 60 years), presence of cholangitis, history of endoscopic papillosphincterotomy (EPST), history of cholecystectomy, blood total bilirubin level (normal or increased), CBD size (≤ 10 or > 10 mm), main pancreatic duct obstruction, proximal level of biliary obstruction. Risk scores were assigned for each factor with the respective β-coefficient. Results. PEP developed in 26 cases (5%). A scoring system was constructed with clinical variables with value module of β coefficients ≥ 0.5: History of EPST (score –1), cholangitis (score –18), proximal obstruction (score 1), pancreatic duct obstruction (score –18), CBD ≤ 10 mm (score 1), normal bilirubin level (score 1). The score value of 1 and more was selected as the cut-off with sensitivity 0.89 and specificity – 0.61. The positive predictive value was 0.1, the negative predictive value was 0.995. The overall accuracy was 61.9%.Conclusion. Our scoring system may be useful in clinical practice to predict safety of ERCP in relation to PEP with a score ≤ 0 or to select high-risk patients with a score ≥ 1.