Predictors and outcomes of moderately severe acute pancreatitis - Evidence to reclassify.

BACKGROUND/OBJECTIVES: After the creation of the moderately severe acute pancreatitis (MSAP) category in the Revised Atlanta Classification in 2012, predictors to identify these patients early have not been identified. The MSAP category includes patients with (peri)pancreatic necrosis, fluid collections, and transient organ failure in the same category. However, these outcomes have not been studied to determine whether they result in similar outcomes to merit inclusion in the same severity.

METHODS

Retrospective, review of 514 consecutive, direct admissions for acute pancreatitis from 2010 to 2013. Multivariate logistic regression identified predictors of MSAP.

RESULTS

Persistent SIRS was the best prognostic marker of MSAP with AUC 0.72. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy for persistent SIRS to predict MSAP are: 55%, 88%, 40%, 93%, and 84%. Patients with necrosis had significantly longer length of stay (LOS) (p = 0.0001) and higher rates of ICU admission (p = 0.02) compared with patients with transient organ failure. Compared to those with acute fluid collections, patients with necrosis had longer LOS (p < 0.0001), higher rates of ICU admission (p = 0.0005), required more interventions (p = 0.001), and demonstrated higher mortality (0.003).

DISCUSSION

Moderately severe pancreatitis can be distinguished from mild pancreatitis on the basis of persistent SIRS but cannot be accurately distinguished from severe pancreatitis in the first 48 h (Peri)pancreatic necrosis demonstrates significantly more morbidity compared to the other components of MSAP of fluid collections and transient organ failure.