Logistic Regression Analysis Of Prognostic Factors In Severe Acute Pancreatitis

Objective To determine the factors correlated with disease severity, mortality and morbidity in patients with severe acute pancreatitis(SAP).Methods Thirty patients with SAP admitted to the intensive care unit (ICU)of Qilu Hospital from January 2004 to October 2008 were analysed retrospectively. The patients were divided into survival group and death group according to their outcomes.25 Data were collected including course, age, gender, etiology, ascites, mean arterial pressure, urine volume in the first 24 hours after admission, Ranson score, APACHE II score in the first 48 hours after admission, Balthazar CT index, and other laboratory parameters performed within 48 hours after the initial admission.Results Significant differences between two groups were found in age, serum albumin(ALB), blood urea nitrogen(BUN), blood creatinine, hydrogen ion concentration(pH), Ranson score and APACHE II score. Logistic regression analysis indicated that the older age, higher blood urea nitrogen(BUN), lower serum albumin(ALB) correlate with SAP mortality. The area under the receiver operating characteristic curve value(AUC) was 0.9850. The sensitivity, specificity, positive predictive value and negative predictive value of the equation in predicting outcome were 0.90,0.95,0.90 and 0.95 respectively.Conclusions The results of our study show that age, ALB, BUN, blood creatinine, pH, Ranson score and APACHEâ…¡score are beneficial to predicting the prognosis of SAP. The older age, higher blood urea nitrogen(BUN),lower serum albumin(ALB) could predict early death for patients with SAP. The equation has important value for forecasting prognosis of severe acute pancreatitis, guiding its treatment and reducing complications and mortality.