Medical Imaging Department, University and Polytechnic Hospital la Fe, Valencia, Spain.
Department of Gastroenterology, University and Polytechnic Hospital la Fe, Valencia, Spain.
Eur J Radiol. 2020 Sep;130:109193. doi: 10.1016/j.ejrad.2020.109193. Epub 2020 Jul 24.
Determine if the pancreatic necrosis volume (PNV) in computed tomography scan (CT) is a useful marker to predict the severity of acute pancreatitis (AP) comparing its predictive value with current clinical scoring systems.
This retrospective study was conducted in a tertiary hospital, including patients hospitalized with AP during the period of 24 months. Demographic, clinical data, length of hospital stay and analytical parameters were collected from the hospital clinical information digital systems. Other information on the severity of the disease was also reviewed, including BISAP score, organ failure (OF) or admission to the ICU, as well as, complications during hospitalization as infected necrotic collections, surgical procedure or death. The quantification of the necrosis volume, CT severity index and Balthazar score were assessed in the CT studies. ROC curves were carried to compare the correlation between different scoring systems and the acute complications.
This study included 163 patients with AP. The calculated average value of PNV in the CT studies was 242 cc (0-1575 cc). PNV showed lineal correlation with hospital stay (Pearson 0.696) and statistically significant association with acute complications as OF, multiple organ failure, infection, need of treatment or hospitalization at ICU (P < 0.05). The optimal cut-off value for predicting complications of necrosis as infections or need of surgery treatments was 75 cc. Sensibility and specificity were 100 % and 78 %, respectively. ROC curves showed that PNV was the best radiological finding correlated with AP complications.
Necrosis volume is a radiological biomarker highly correlated with AP complications.
通过比较其预测价值与当前临床评分系统,确定 CT 扫描中胰腺坏死体积(PNV)是否是预测急性胰腺炎(AP)严重程度的有用标志物。
这项回顾性研究在一家三级医院进行,纳入了 24 个月期间因 AP 住院的患者。从医院临床信息数字化系统中收集人口统计学、临床数据、住院时间和分析参数。还回顾了其他疾病严重程度的信息,包括 BISAP 评分、器官衰竭(OF)或入住 ICU 以及住院期间的并发症,如感染性坏死性积聚、手术或死亡。在 CT 研究中评估了坏死体积、CT 严重程度指数和 Balthazar 评分的定量。进行 ROC 曲线以比较不同评分系统与急性并发症之间的相关性。
这项研究纳入了 163 例 AP 患者。CT 研究中计算的平均 PNV 值为 242cc(0-1575cc)。PNV 与住院时间呈线性相关(Pearson 0.696),与 OF、多器官衰竭、感染、需要在 ICU 治疗或住院等急性并发症有统计学显著关联(P<0.05)。预测感染或需要手术治疗的坏死并发症的最佳截断值为 75cc。敏感性和特异性分别为 100%和 78%。ROC 曲线显示 PNV 是与 AP 并发症相关性最强的影像学发现。
坏死体积是与 AP 并发症高度相关的影像学生物标志物。