Purpose: In this study, we aimed to evaluate the predictive value of Stewart-derived parameters for the development of severe type of acute pancreatitis (AP) and for AP-related mortality.
Methods: We studied 186 patients admitted to the hospital with AP. We performed blood gas and biochemical analysis for each patient on admission. We calculated multiple metrics according to the Stewart's acid-base theory and assessed their accuracy as predictors of AP severity and mortality.
Results: Of the 186 patients presenting with AP, 85 (45.7%) developed severe AP and 33 (17.7%) died during hospitalization. Patients with severe AP had significantly higher median strong ion gap (SIG) than did patients with mild or moderate AP (7.88 vs 2.11 mEq/L, P< .001). In multivariate logistic regression analysis, SIG had an odds ratio (OR) of 1.56 (P< .001). In addition, SIG had good predictive power for mortality (OR, 1.26; P= .014) as well as acute kidney injury (OR, 1.34; P< .001).
Conclusions: In a cohort of patients with AP, SIG was a strong independent predictor of severity and mortality. Besides, SIG might also be an early marker for acute kidney injury in AP patients. Additional research is needed to identify the nature of the unmeasured anions responsible for such findings.
Keywords: Acute pancreatitis; Albumin; Anion gap; Lactate; Outcome prediction; Strong ion gap.
Copyright © 2016 Elsevier Inc. All rights reserved.