Revised Marshall Score: A New Approach to Stratifying the Severity of Acute Pancreatitis

Modified Marshall Score is one of the severity scores for acute pancreatitis (AP) and is included in the Revised Atlanta Classification, but given its utilization of a set serum creatinine level (sCr), it may misclassify stable patients with chronic kidney disease (CKD) to a more severe class just due to their elevated sCr.

Our study aims to evaluate the role of CKD in AP and the possibility of utilizing acute kidney injury (AKI) into developing a new scoring system.

We retrospectively reviewed the electronic medical records of three hundred consecutive patients who were diagnosed with AP during hospitalization. Multiple demographic variables and clinical course indices were collected. Univariate logistic regression was then applied to predict mortality and ICU admission. Finally, receiver operating curve was utilized to compare original versus New Revised Marshall Score.

Two hundred and eight-four (284) patients had a definitive diagnosis of AP. When comparing patients who had AKI on admission to those without AKI, the AKI group showed statistically significant higher mortality rate (5.6% vs. 1.1%, p = 0.04). Finally, we substituted the renal part of Marshall Score with our AKIN and we plotted the New "Revised" Marshall Score, which showed a higher AUROC compared to the original modified version (C-statistics 0.93 vs. 0.89, p < 0.05).

We found that AKI predicts mortality and outperforms the use of a fixed sCr value alone. The use of our New Revised Marshall Score can accurately classify AP severity, avoiding misclassification of AP severity and providing better patient care.