Objective To introduce a novel modified CT severity index based on the assessment of extrapancreatic inflammation and pancreatic necrosis on CT index (EPIPN) and to evaluate its effect in predicting the severity and prognosis of acute pancreatitis. Methods Seventy-seven consecutive patients diagnosed as acute pancreatitis (AP) from August 2006 to December 2007 were retrospectively analyzed. The clinical data included age, sex, cause, the C-reactive protein(CRP) level with in 72 hours of onset of symptom, Ranson signs, the APACHE I1 score, the disappearing time of the abdominal pain, the presence of organ failure, the length of hospital stay, etc. All patients underwent contrast-enhanced multisection CT scan after admission of 48-72 hours. The CT severity index (CTSI) and EPIPN scores were obtained. The severity of pancreatitis for each patient was then categorized as severe if CTSI≥7 or EPIPN>5. The diagnostic value of EPIPN in predicting the severe acute pancreatitis (SAP) was compared with that of CTSI using ROC curve. The correlation of EPIPN or CTSI with clinical coutcome was conducted. Results Of 77 patients, 34 were males and 43 were females with mean age of 51.79 years (age range 22-92 years). The causes of AP were gallstones (63 cases), hyperlipemia (6 cases), alcohol (1 case) and idiopathic (7 cases). Organ system failure was present in 14 (18.2%) of the 77 patients. The area under the ROC curve of CTSI in predicting the SAP was 0.72 (95% CI: 0.59-0.88) with sensitivity of 80.4% and specificity of 55% when CTSI≥7, and that in EPIPN was 0.82 (95% CI: 0.73-0.91) with sensitivity of 91.3% and specificity of 63% when EPIPN >5. EPIPN was well correlated with hospital stay, APACHE Ⅱ score and CRP levels. Conclusions The EPIPN allows accurate estimation of disease severity and prognosis in AP patients. The diagnostic effect of EPIPN for predicting SAP is superior to CTSI. The EPIPN index is-both convenient and practical, and has clinical value. Key words: Pancreatitis,acute necrotizing; Severity of disease index; Prognosis
Read full abstract