Risk factors for recurrent acute alcohol-associated pancreatitis: A prospective analysis

Abstract

Objective. In an earlier retrospective study we showed that 46% of patients with acute alcoholic pancreatitis had recurrent attacks within 10–20 years, about 30% having a recurrence during the first 3 years. The aim of this prospective follow-up study was to determine the risk factors associated with recurrences. Material and methods. Sixty-eight patients, (59 M, 9 F, mean age 46 years, range 25–71 years) who survived their first acute alcohol-induced pancreatitis from January 2001 to January 2004 volunteered to participate in the study. The diagnostic criteria for acute pancreatitis were epigastric pain, serum amylase >3 times the upper normal range, elevated serum C-reactive protein (CRP), and signs of acute pancreatitis in imaging. Other etiologies were excluded. Alcohol consumption and dependency were detected by the Alcohol Use Disorders Identification Test (AUDIT) and the Short Alcohol Dependence Data (SADD), respectively, and by attempting to evaluate recent use in grams of pure alcohol. Social and demographic data of the patients, smoking, body mass index, and the severity of the pancreatitis were recorded. Serum and fecal markers of the endocrine and exocrine function and secretin-stimulated MRCP were studied. The patients were followed for a median 38 (25–61) months for recurrences, and at the 2-year time-point had a follow-up visit to investigate any changes in alcohol consumption. Results. Seventeen (25%) patients had recurrences of acute alcoholic pancreatitis during the follow-up. Pre-illness alcohol consumption, the severity of the pancreatitis, patient's social or demographic data, pancreatic function tests or morphologic changes in MRCP, or smoking did not correlate with recurrence. None of the 13 patients with consistent total abstinence from alcohol at 2 years developed recurrent pancreatitis compared with 17 out of 51 (33%) patients with at least some alcohol consumption (p=0.02). Use of other sedatives than alcohol before the first attack of pancreatitis was an independent risk factor associated with recurrence (HR = 6.95, 95% CI 2.45–19.72, p<0.001). A lower reduction in dependency on alcohol (less decreased SADD points) during 2 years was associated with a higher recurrence rate (HR = 0.921/each reduced point, 95% CI 0.872–0.974, p=0.004). Conclusions. Contrary to chronic pancreatitis, smoking was not found to be a risk factor for recurrent episodes after the first attack of acute alcoholic pancreatitis. Abstinence from alcohol protects against recurrent pancreatitis. Patients who developed recurrent acute pancreatitis had increased dependency on alcohol, demonstrated by the use of other sedatives in addition to alcohol and supported by the less decreased dependency during the follow-up.