摘要目的:探讨计算CT严重指数（CTSI）和改良CT严重指数（MCTSI）对修订版亚特兰大分类（RAC）下急性胰腺炎（AP）严重程度的评估作用，及其预测临床预后的价值。方法:基于前瞻性录入的AP数据库，回顾性筛选2012年1月至2020年12月南昌大学第一附属医院消化内科连续收治的成人AP住院患者临床资料，由两名影像科医生独立评估影像学资料录入数据库中计算CTSI和MCTSI评分，分析其与RAC严重程度分级的差异和临床预后的关系；以急性生理和慢性健康评分Ⅱ（APACHE Ⅱ）为对照，采用受试者工作特征曲线评价CTSI和MCTSI对持续器官衰竭和感染性胰腺坏死（IPN）的预测价值。结果:研究共纳入2 612例AP患者，年龄（50±15）岁，其中男1 547例（59.2%），女1 065 例（40.8%）。按RAC标准分类，AP分为轻症胰腺炎（MAP）699例（26.8%）、中度重症胰腺炎（MSAP）1 098例（42.0%）、重症胰腺炎（SAP）815例（31.2%）；MCTSI对AP严重程度的判定与RAC相近，分别为MAP 668例（25.6%）、MSAP 1 207例（46.2%）、SAP 737例（28.2%），而CTSI判定的SAP患者较少（400例，15.3%）。CTSI和MCTSI评分判定的AP严重程度与临床预后均相关（ r=0.06~0.43，均 P<0.05）。与APACHE Ⅱ评分比较，CTSI预测IPN的曲线下面积（AUC）最高（AUC=0.85，95% CI：0.83~0.87），其次是MCTSI（AUC=0.82，95% CI：0.80~0.85）；APACHE Ⅱ预测持续性器官衰竭的准确性高于CTSI和MCTSI评分，其AUC分别为0.73（95% CI：0.71~0.75）、0.72（95% CI：0.70~0.74）和0.72（95% CI：0.70~0.74）。 结论:MCTSI判定的AP严重程度与RAC一致，CTSI判定的SAP患者较RAC少。CTSI和MCTSI均与临床预后显著相关；CTSI和MCTSI预测IPN的准确性较高，但预测持续性器官衰竭的准确性低于APACHE Ⅱ。

abstractsObjectives:To explore the role of computed tomography (CT) severity index (CTSI) and modified CT severity index (MCTSI) in assessing the severity of acute pancreatitis (AP) under the revised Atlanta classification (RAC) and predicting the clinical prognosis.Methods:Based on the prospectively entered AP database, the clinical data of consecutive adult AP inpatients admitted to the Department of Gastroenterology of the First Affiliated Hospital of Nanchang University from January 2012 to December 2020 were retrospectively screened. The imaging data were independently evaluated by two radiologists and entered to the database to calculate the CTSI and MCTSI scores. Their relationship with the difference of RAC severity grade and clinical prognosis was analyzed. Compared with Acute Physiology and Chronic Health Assessment Ⅱ (APACHE Ⅱ) score, the receiver operating characteristic curve was used to evaluate the predictive value of CTSI and MCTSI scores for persistent organ failure and infectious pancreatic necrosis (IPN).Results:A total of 2 612 patients with AP, aged (50±15) years, were included in the study, including 1 547 males (59.2%) and 1 065 females (40.8%). According to RAC standard, AP was divided into 699 cases (26.8%) of mild pancreatitis (MAP), 1 098 cases (42.0%) of moderately severe pancreatitis (MSAP), and 815 cases (31.2%) of severe pancreatitis (SAP). MCTSI judged AP severity similarly to RAC, with 668 cases of MAP (25.6%), 1 207 cases of MSAP (46.2%) and 737 cases of SAP (28.2%), while CTSI judged SAP patients less(400 cases, 15.3%). The severity of AP determined by CTSI and MCTSI scores was significantly correlated with clinical prognosis ( r=0.06-0.43, all P<0.05). Compared with APACHE Ⅱ score, CTSI had the highest area under the curve (AUC) for predicting IPN (AUC=0.85, 95% CI: 0.83-0.87), followed by MCTSI (AUC=0.82, 95% CI: 0.80-0.85). APACHE Ⅱ was more accurate in predicting persistent organ failure than CTSI and MCTSI scores,with AUC of 0.73 (95% CI: 0.71-0.75) , 0.72 (95% CI: 0.70-0.74) and 0.72 (95% CI: 0.70-0.74), respectively. Conclusions:AP severity judged by MCTSI is consistent with RAC, and SAP patients judged by CTSI are less than RAC. CTSI and MCTSI are significantly correlated with clinical prognosis. CTSI and MCTSI have higher accuracy in predicting IPN, but lower accuracy in predicting persistent organ failure than APACHE Ⅱ.