Prediction of outcome in acute pancreatitis:a comparative study of APACHEII. Ranson and Balthazar computed tomography scoring systems
Li Zhao
Abstract
Li Zhao
Abstract
Objective To evaluate the prognostic ability of three different scoring systems (four group scores), which are very useful in clinic, and the relations between prognostic variables and scoring systems. Methods 201 patients with acute severe pancreatitis, who were admitted into Changhai Hospital from January 1993 to January 2004, were studied retrospectively. Data pertinent to scoring systems were recorded 24 hours (APACHEⅡscore). 48 hours (APACHEⅡand Ranson scores) and 72 hours (Balthazar computed tomography severity index) after admission. The prognostic variables include: pancreatic necrosis, local complications, infection, and development of organ failure, surgery drainage, death, and the days of food prohibition and in hospital. Statistical analysis was performed by using receiver operating characteristic curves and by logistic regression for all eight outcome variables. Results Through compared AUC. in organ failure and days of food prohibition prediction. APACHEⅡ(48 h) score was superior. In predicting local complications, the days of food prohibition and in hospital, Ranson score was proper to be recommended. CT score performed obviously better than other scores in pancreatic necrosis and surgery drainage prediction. Conclusions For every eight outcome variables, the prognostic ability of different scoring systems was not same. Several scoring systems should combine to predict the outcome of severe acute pancreatitis.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Objective To evaluate the prognostic ability of three different scoring systems (four group scores), which are very useful in clinic, and the relations between prognostic variables and scoring systems. Methods 201 patients with acute severe pancreatitis, who were admitted into Changhai Hospital from January 1993 to January 2004, were studied retrospectively. Data pertinent to scoring systems were recorded 24 hours (APACHEⅡscore). 48 hours (APACHEⅡand Ranson scores) and 72 hours (Balthazar computed tomography severity index) after admission. The prognostic variables include: pancreatic necrosis, local complications, infection, and development of organ failure, surgery drainage, death, and the days of food prohibition and in hospital. Statistical analysis was performed by using receiver operating characteristic curves and by logistic regression for all eight outcome variables. Results Through compared AUC. in organ failure and days of food prohibition prediction. APACHEⅡ(48 h) score was superior. In predicting local complications, the days of food prohibition and in hospital, Ranson score was proper to be recommended. CT score performed obviously better than other scores in pancreatic necrosis and surgery drainage prediction. Conclusions For every eight outcome variables, the prognostic ability of different scoring systems was not same. Several scoring systems should combine to predict the outcome of severe acute pancreatitis.
Key concepts: Medicine, Acute pancreatitis, Logistic regression, Receiver operating characteristic, Scoring system, Pancreatitis, Computed tomography, Outcome (game theory)