Objective To analyze the clinical features and risk factors of recurrent acute pancreatitis (RAP). Methods From September 2012 to September 2014, the clinical data of 411 patients with primary acute pancreatitis (AP) were collected. From March to June 2016, patients were followed up. The clinical features of patients with RAP were analyzed. Univariate and multivariate regression analysis were performed to analyze the risk factors of RAP. Results Among the 411 patients with AP, those caused by biliary disease, hyperlipidemia, alcohol, other known causes and idiopathic AP were 265 cases (64.5%), 61 cases (14.8%), 19 cases (4.6%), 21 cases (5.1%) and 45 cases (10.9%), respectively. In two weeks of AP onset, the recurrent rate of biliary AP in cholecystectomy group was 7.1%(5/70), which was lower than that of non-cholecystectomy group (30.2%, 42/139), and the difference was statistically significant (χ2=14.218, P<0.01). The results of univariate regression analysis suggested that gender, body mass index (BMI), complicated with diabetes, etiology, history of smoking, history of drinking and pancreatic necrosis were correlated with RAP (all P<0.05). The results of multivariate regression analysis indicated that complicated with diabetes (odd ratios (OR)=3.417, 95% confidence interval (CI) 1.979 to 5.900, P<0.01), hyperlipidemic pancreatitis (OR=2.247, 95%CI 1.077 to 4.688, P=0.023), history of smoking (OR=4.023, 95%CI 2.377 to 6.809, P<0.01), complicated with pancreatic necrosis (OR=3.312, 95%CI 1.675 to 6.546, P<0.01) were independent risk factors of RAP. Conclusions Hypertriglyceridemia, smoking, complicated with pancreatic necrosis and diabetes are independent risk factors of RAP. Patients with biliary AP should receive cholecystectomy as early as possible, which could reduce RAP. Key words: Pancreatitis, recurrent acute; Risk factors; Clinical features