Liu Dingxia, Chen Jiejun, Zhang Yunfei, Gu Yajia, Yao Xiuzhong
Shanghai Institute of Medical Imaging, No 130, Dongan Rd, Xuhui District, Shanghai200032, PR China (D.L., Y.Z., Y.G., X.Y.); Department of Radiology, Zhongshan Hospital, Fudan University, No 138, Fenglin Rd, Xuhui District, Shanghai 200032, PR China (J.C., X.Y.).
Shanghai Institute of Medical Imaging, No 130, Dongan Rd, Xuhui District, Shanghai 200032, PR China (J.C., X.Y.).
Acad Radiol. 2025 Aug;32(8):4555-4563. doi: 10.1016/j.acra.2025.03.028. Epub 2025 Apr 2.
To investigate the magnetic resonance elastography (MRE)-derived pancreatic stiffness for predicting the occurrence of clinically relative postoperative pancreatic fistula (CR-POPF) in patients with partial pancreatectomy, and establish a predictive model for POPF before surgery.
Pancreatic stiffness reflects fibrosis and fat infiltration, which are associated with CR-POPF. But preoperative prediction remains a challenge. MRE was proven to evaluate pancreatic stiffness accurately, potentially being a predictive imaging biomarker of POPF.
This prospective study included adult patients who underwent magnetic resonance imaging with MRE sequence and subsequent partial pancreatectomy between August 2021 and December 2023. The relationships of MRE stiffness and main pancreatic duct diameter (MPD) with the risk of POPF were analyzed using logistic regression. Independent risk factors were identified to construct the nomogram prediction model. The predictive performance of each parameter and the model was conducted by calculating the area under the ROC curve (AUC).
A total of 73 patients (age 58.99±12.55 years; 30 pancreatoduodenectomy and 43 distal pancreatectomy) were enrolled, among whom 15 developed CR-POPF and 58 did not. After conducting uni- and multivariate logistic regression analyses, high BMI was found to be an independent risk factor for the occurrence of POPF (OR=2.916, 95% CI: 1.472-9.394, P=0.02), while high pancreatic MRE stiffness (OR=0.0633, 95% CI: 0.0022-0.5273, P=0.04) and large MPD (OR=0.0728, 95% CI: 0.003-0.5165, P=0.04) were independent protective factors. A preoperative prediction model for POPF was constructed by combining the three indicators, which has excellent predictive performance with an AUC of 0.97.
MRE can quantitatively evaluate the mechanical property of pancreas, which is a reliable indicator for predicting the risk of POPF. The POPF prediction model established by combining BMI, pancreatic stiffness value, and MPD has promising clinical application prospects.
探讨磁共振弹性成像（MRE）得出的胰腺硬度对预测部分胰腺切除患者临床相关术后胰瘘（CR-POPF）发生情况的作用，并建立术前预测胰瘘的模型。
胰腺硬度反映纤维化和脂肪浸润情况，这与CR-POPF相关。但术前预测仍是一项挑战。MRE已被证明能准确评估胰腺硬度，可能成为胰瘘的预测性影像生物标志物。
这项前瞻性研究纳入了2021年8月至2023年12月期间接受磁共振成像MRE序列检查并随后接受部分胰腺切除术的成年患者。使用逻辑回归分析MRE硬度和主胰管直径（MPD）与胰瘘风险的关系。确定独立危险因素以构建列线图预测模型。通过计算ROC曲线下面积（AUC）评估各参数及模型的预测性能。
共纳入73例患者（年龄58.99±12.55岁；30例行胰十二指肠切除术，43例行远端胰腺切除术），其中15例发生CR-POPF，58例未发生。进行单因素和多因素逻辑回归分析后，发现高体重指数是胰瘘发生的独立危险因素（OR=2.916，95%CI：1.472-9.394，P=0.02），而高胰腺MRE硬度（OR=0.0633，95%CI：0.0022-0.5273，P=0.04）和大MPD（OR=0.0728，95%CI：0.003-0.5165，P=0.04）是独立保护因素。结合这三个指标构建了术前胰瘘预测模型，其预测性能优异，AUC为0.97。
MRE可定量评估胰腺的力学特性，是预测胰瘘风险的可靠指标。结合体重指数、胰腺硬度值和MPD建立的胰瘘预测模型具有良好的临床应用前景。
