Abstract

Objective To investigate the risk factors of pancreatic pseudocysts (PPC) in patients with acute pancreatitis (AP) in a retrospective cohort study. Methods 460 AP patients with complete follow-up data admitted in Affiliated Hospital of Qingdao University from January 2004 to March 2012 were retrospectively analyzed, who were divided into PPC group and control group. Age, gender, body mass index (BMI), history of diabetes, etiology, the presence of ascites and hydrothorax, the presence of abdominal mass, the presence of acute fluid collection, APACHEⅡ score at 48 h admission, CT severity index (CTSI), serum albumin, amylase, LDH, ALP, BUN, Cr, TG, TB, conjugated bilirubin, CRP, serum calcium and other laboratory markers were recorded. Univariate logistic regression analysis was used to select the factors that were statistically different between two groups, and multivariate logistic regression analysis was performed to determine the independent risk factors for AP complicated with PPC. Results 143(31.1%) of 460 AP patients developed PPC. On univariate analysis, a total of 11 factors including male sex, BMI ≥28 kg/m2, history of diabetes, alcoholic pancreatitis, ascites, pleural effusion, palpable abdominal mass, acute fluid collections, APACHEⅡ score, CTSI≥7 and serum albumin were statistically different between two groups. On multiple logistic regression analysis, it was shown that male sex (OR 3.23, 95%CI 1.560~6.301, P=0.03), history of diabetes (OR 2.23, 95% CI 1.021~3.920, P=0.04), ascites (OR 1.62, 95% CI 0.652~2.432, P=0.01), pleural effusion (OR 2.43, 95% CI 1.201~7.201, P=0.03), a palpable abdominal mass (OR 1.83, 95% CI 0.737~4.320, P<0.001) and CTSI ≥7 (OR 5.12, 95% CI 1.890~14.012, P<0.001) were independent risk factors significantly associated with the PPC formation. Conclusions The male sex, diabetic history, ascites, pleural effusion, palpable abdominal mass and high CTSI score were the independent risk factors of PPC formation in AP. Key words: Pancreatitis; Pancreatic pseudocyst; Risk factor; Logistic regression analysis

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call