Early prediction of organ failure under the revised Atlanta classification

This study aimed to compare the ability of conventional laboratory markers and scoring systems to early predict organ failure (OF) and to differentiate between transient and persistent OF in patients with acute pancreatitis (AP) using the revised Atlanta classification.

We retrospectively analyzed the medical records of 214 patients with AP between January 2014 and July 2015. The predictive values of laboratory markers were analyzed. The predictive accuracy of individual markers, extrapancreatic inflammation on computed tomography (EPIC), acute physiology and chronic health evaluation II (APACHE II), and bedside index for severity in acute pancreatitis (BISAP) scores were measured using the area under the receiver operating characteristic curve (AUROC).

OF was diagnosed in 32 (15%) patients and persistent OF in 14 (6.5%). There were statistically significant differences between patients with and without OF with respect to white blood cell count, creatinine, blood urea nitrogen, lactate dehydrogenase, C-reactive protein, calcium (Ca), arterial partial pressure of oxygen (PaO2), base excess (BE), APACHE II, BISAP scores, and EPIC scores. Logistic regression analysis identified Ca, PaO2, and BE as independent predictors of OF. Using AUROC, the EPIC score had the highest accuracy for the early prediction of OF, which was 0.82. No significant differences were detected between patients with transient and persistent OF.

Several laboratory markers and score systems were useful for the early prediction of OF in patients with AP, of which Ca, PaO2, and BE had highest predicting value, and EPIC score had the highest accuracy. We could not predict the duration of OF using laboratory markers.