APACHE III Score as a Prognostic Parameter in Acute Pancreatitis

Park, Kyung Nam; Lee, Min Ho; Choi, Ho Soon; Park, Joon Yong; Han, Dong Soo; Sohn, Joo Hyun; Hahm, Joon Soo; Cho, Yun Ju; Mun, Gwang Ho; Yun, Hyung Ran; Yang, Seok Chul; Khee, Chun Suhk

Abstract

BACKGROUND AND AIMS: The various prognostic parameters for predicting the severity and prognosis of acute pancreatitis were used, but these were unfit for clinical utilization. We evaluated a clinical utility of the APACHE III score as a prognostic parameter of acute pancreatitis.
METHODS
We conducted a retrospeetive review of 40 patients with acute pancreatitis. We compared APACHE III score with Ranson's score, Glasgow's score, CT grading, the severity of complication, and hospital day.
RESULTS
The sensitivity of these prognostic parameters for predicting the severity of acute pancreatitis was 69.57% in APACHE III score (>23), 47.83% in Glasgow's score (>3), 39.13% in Ranson's score (>3) and 73.91% in CT pading (>3). The specificity was 64.71% in APACHE III score, 76.41% in Glasgow's score, 64.71% in Ranson's score and 41.18% in CT grading. The positive predictive value of APACHE III score in predicting the severity of acute pancreatitis was 72.73%. We found a good correlation between the severity of complication and APACHE III seore (r =0.607). The severity of complication is poorly correlated with other parameters: Glasgow's score (r=0.387), Ranson's score (r=0.402) and CT grading (r =0.273). There was not statistically significant difference in the APACHE III score (32.41 vs. 22.94, p=0.089) between the groups of alcohol (n =22 cases) and biliary (n =16 cases) associated pancreatitis.
CONCLUSIONS
The APACHE III score showed a higher sensitivity and specificity than the other prognostic parameters and excellent correlation with the severity of complication and clinically initial systemic states in the acute pancreatitis.