Pancreatic exocrine insufficiency following acute pancreatitis: Systematic review and study level meta-analysis

Abstract

ObjectivesThis study systematically explores the prevalence of pancreatic exocrine insufficiency (PEI) after acute pancreatitis in different subgroups of etiology (biliary/alcoholic/other), disease severity and follow-up time (<12, 12–36 and > 36 months after index admission). MethodsPubMed and EMBASE databases were searched, 32 studies were included in this study level meta-analysis. ResultsIn a total of 1495 patients with acute pancreatitis, tested at a mean of 36 months after index admission, the pooled prevalence of PEI was 27.1% (95%-confidence interval [CI]: 20.3%–35.1%). Patients from seven studies (n = 194) underwent direct tests with pooled prevalence of 41.7% [18.5%–69.2%]. Patients from 26 studies (n = 1305) underwent indirect tests with pooled prevalence of 24.4% [18.3%–31.8%]. In subgroup analyses on patients that underwent fecal elastase-1 tests, PEI occurred more often in alcoholic pancreatitis (22.7% [16.6%–30.1%]) than in biliary pancreatitis (10.2% [6.2%–16.4%]) or other etiology (13.4% [7.7%–22.4%]; P = 0.02). Pooled prevalence of PEI after mild and severe pancreatitis was 19.4% [8.6%–38.2%] and 33.4% [22.6%–46.3%] respectively in studies using fecal elaste-1 tests (P = 0.049). Similar results were seen in patients without (18.9% [9.3%–34.6%]) and with necrotizing pancreatitis (32.0% [18.2%–49.8%]; P = 0.053). Over time, the prevalence of PEI decreased in patients who underwent the fecal elastase-1 test and increased in patients who underwent the fecal fat analysis. ConclusionsAfter acute pancreatitis, a quarter of all patients develop PEI during follow-up. Alcoholic etiology and severe and necrotizing pancreatitis are associated with higher risk of PEI. The prevalence of PEI may change as time of follow-up increases.