The challenge of prognostic markers in acute pancreatitis: internist’s point of view

Acute pancreatitis, the most frequent hospitalization reason in internal medicine ward among gastrointestinal diseases, is burdened by high mortality rate. The disease manifests mainly in a mild form, but about 20-30% patients have a severe progress that requires intensive care. Patients presenting with acute pancreatitis should be clinically evaluated for organ failure signs and symptoms. Stratifying patients in the first days from symptoms onset is essential to determine therapy and care setting. The aim of our study is to evaluate prognostic factors for acute pancreatitis patients, hospitalized in internal medicine wards, and moreover, understanding the role of various prognostic scores validated in intensive care setting in predicting in-hospital mortality and/or admission to intensive care unit. We conducted a retrospective study enrolling all patients with diagnosis of acute pancreatitis admitted took an internal medicine ward between January 2013 and May 2019. Adverse outcome was considered in-hospital mortality and/or admission to intensive care unit. In total, 146 patients (137 with positive outcome and 9 with adverse outcome) were enrolled. The median age was (67.89 ± 16.44), with a slight prevalence of male (55.1%) compared to female (44.9%). C protein reactive (p = 0.02), creatinine (p = 0.01), sodium (p = 0.05), and troponin I (p = 0.013) after 48 h were significantly increased in patients with adverse outcome. In our study, progression in SOFA score independently increases the probability of adverse outcome in patients hospitalized with acute pancreatitis. SOFA score > 5 is highly predictive of in-hospital mortality (O.R. 32.00; C.I. 6.73-152.5; p = 0.001) compared to other scores. The use of an easy tool, validated in intensive care setting such as SOFA score, might help to better stratify the risk of in-hospital mortality and/or clinical worsening in patients hospitalized with acute pancreatitis in internal medicine ward.