机构地区:[1]南京大学医学院附属鼓楼医院核医学科,江苏南京210000 [2]南京大学医学院附属鼓楼医院胆胰外科,江苏南京210000
出 处:《中国中西医结合影像学杂志》2022年第3期238-243,共6页Chinese Imaging Journal of Integrated Traditional and Western Medicine
基 金:南京鼓楼医院临床课题面上项目(2021-LCYJ-MS-11)。
摘 要:目的:探讨胰腺导管内乳头状黏液性肿瘤(IPMN)恶变的高危因素,建立术前恶性风险列线图预测模型并予以验证。方法:回顾性分析经术后病理证实的72例IPMN患者的临床与影像学资料。以术后病理为标准,将患者分为良性组和恶性组。采用Logistic回归分析建立IPMN术前恶性风险预测模型,并绘制列线图使模型结果可视化。使用ROC曲线和校正曲线评价模型的准确性和校准度。结果:72例中,恶性占31.9%(23/72)。单因素分析表明,CA19-9>27 U/mL、中性粒细胞淋巴细胞比值(NLR)>1.81、血小板淋巴细胞比值(PLR)>124、肿瘤最大径、胰管扩张和增强壁结节均与IPMN恶变显著相关(均P<0.05)。将单因素筛选出来的特征纳入Logistic多元回归分析进一步明确,CA19-9>27 U/mL(OR=5.37,P=0.063)、PLR>124(OR=10.86,P=0.008)、肿瘤最大径(OR=1.06,P=0.024)、胰管扩张(OR=6.18,P=0.020)和增强壁结节(OR=11.93,P=0.019)是IPMN恶变的独立危险因素。绘制模型的ROC曲线,AUC值为0.928,敏感度100.00%,特异度79.59%。列线图预测模型显示,肿瘤最大径是预测IPMN恶变的最大贡献因子(100分),其次为增强壁结节(50分)、PLR(49分)、胰管扩张(37分)和CA19-9(34分),同时校正曲线显示该列线图预测模型具有较好的校准度。结论:CA19-9>27 U/mL、PLR>124、肿瘤最大径、胰管扩张和增强壁结节是恶性IPMN发生的独立危险因素,基于此建立的IPMN恶性风险列线图预测模型具有较高的准确率和校准度,可为临床决策提供参考。Objective:To explore the high-risk factors for malignant transformation of pancreatic intraductal papillary mucinous neoplasm(IPMN)and to develop and validate a preoperative risk nomogram prediction model.Methods:The clinical and imaging data of 72 patients with IPMN confirmed by postoperative pathology were analyzed retrospectively.The patients were divided into benign and malignant groups using postoperative pathology as the criterion.The high-risk factors associated with malignant IPMN were explored,logistic regression was used to establish a prediction model for the risk of IPMN malignancy,and nomogram were drawn to visualize the model results.The accuracy and calibration degree of the model were evaluated by ROC curve and calibration curve.Results:Of the 72 cases,31.9%(23/72)were malignant.Univariate analysis showed that CA19-9,platelet-to-lymphocyte ratio(NLR),platelet-to-lymphocyte ratio(PLR),maximum tumor diameter,pancreatic duct dilatation and enhanced mural nodules were significantly associated with IPMN malignancy,then the screened characteristics by univariate were included in multiple logistic regression analysis to further identify,which showed that CA19-9>27 U/mL(OR=5.37,P=0.063),PLR>124(OR=10.86,P=0.008),maximum tumor diameter(OR=1.06,P=0.024),pancreatic duct dilatation(OR=6.18,P=0.020)and enhanced mural nodules(OR=11.93,P=0.019)were independent risk factors for the development of malignant IPMN,and the ROC curve of the model was plotted with an AUC value of 0.928 with the sensitivity of 100%and specificity of 79.59%.The nomogram prediction model showed that the maximum tumor diameter was the largest contributor(100 points)in predicting IPMN malignancy,followed by enhancement wall nodule(50 points),PLR(49 points),pancreatic duct dilatation(37 points)and CA19-9(34 points),and its calibration curve showed that it had a good degree of calibration.Conclusions:CA19-9>27 U/mL,PLR>124,maximum tumor diameter,pancreatic duct dilatation and enhanced mural nodules are independent risk factors for the develop
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...