| 方垂宝,刘永凡,朱驿,等.胰腺囊性肿瘤潜在恶性风险相关危险因素分析[J].安徽医药,2026,30(10):2028-2033. |
| 胰腺囊性肿瘤潜在恶性风险相关危险因素分析 |
| Risk factors associated with the potential malignancy of pancreatic cystic neoplasms |
| |
| DOI：10.3969/j.issn.1009-6469.2026.10.023 |
| 中文关键词: 胰腺囊性肿瘤 恶变风险 预测模型 良性肿瘤 恶性肿瘤 糖类抗原 19-9 |
| 英文关键词: Pancreatic cystic neoplasms Malignancy risk Prediction model Benign tumor Malignant tumor Carbohydrate antigen 19-9 |
| 基金项目:安徽省高校科研项目计划项目( 2024AH050797)；安徽医科大学校科研基金项目( 2022xkj182)；安徽医科大学第二附属医院转化医学研究科研基金项目( 2022ZHYJ06) |
|
| 摘要点击次数: 110 |
| 全文下载次数: 21 |
| 中文摘要: |
| 目的针对有潜在恶性转化风险的胰腺囊性肿瘤( PCN)病人，术前深入剖析恶变风险危险因素，旨在提供评估手段，精确预测其良恶性转化风险。方法回顾性收集 2018年 1月至 2023年 9月安徽医科大学第一附属医院、安徽医科大学第二附属医院共 145例确诊胰腺囊性肿瘤( PCN)病人的临床数据，根据术后病理检查结果分为低、高风险组，并基于导管内乳头状黏液性肿瘤( IPMN)、黏液性囊性肿瘤( MCN)、浆液性囊性肿瘤( SCN)3种亚型构建 PCN良恶性风险预测模型。结果单因素回归分析揭示性别、临床症状、糖类抗原 19-9(CA19-9)、中性粒细胞与淋巴细胞比值( NLR)主胰管是否扩张、肿瘤位置、肿瘤长径、吸烟史、边界清晰度及强化壁结节与 PCN恶性转化风险相关，其中低风险组肿瘤直径为(、34.72±12.35)mm，高风险组肿瘤长径为(41.26±15.78)mm，组间差异有统计学意义。多因素回归分析确定 CA19-9升高、有临床症状、肿瘤位于头颈部及有吸烟史为独立危险因素。基于此构建模型，受试者操作特征曲线( ROC曲线)的曲线下面积( AUC)为 0.83［95%CI:(0.75，0.90)］，计算出最佳临界值为 0.51，此时模型灵敏度 64.2%，特异度 87.1%，总准确率 77.2%。结论 CA19-9水平升高、有临床症状、肿瘤位于头颈部、有吸烟史提示胰腺囊性肿瘤恶性转化风险升高。基于这些关键要素，构建了一个高准确性的预测模型，有望成为未来临床实践中评估 PCN性质、指导个性化治疗的重要工具。 |
| 英文摘要: |
| Objective To conduct an in-depth preoperative analysis of risk factors associated with the potential for malignant transfor-mation in patients with pancreatic cystic neoplasms (PCN) that pose a risk of malignancy, aiming to provide an assessment tool for accu-rately predicting the risk of benign-to-malignant transformation.Methods Clinical data were retrospectively collected from 145 pa-tients diagnosed with PCN at the First and Second Affiliated Hospitals of Anhui Medical University between January 2018 and Septem-ber 2023. Patients were classified into low-and high-risk groups based on postoperative pathological examination results. A predictionmodel for benign and malignant risks of PCN was developed for patients with the three subtypes: intraductal papillary mucinous neo-plasms (IPMN), mucinous cystic neoplasms (MCN), and serous cystic neoplasms (SCN). Results Univariate regression analysis re-vealed that gender, clinical symptoms, carbohydrate antigen19-9 (CA19-9) levels, neutrophil-to-lymphocyte ratio (NLR), main pancreat-ic duct dilation, tumor location, tumor size, smoking history, boundary clarity, and enhanced mural nodules were associated with therisk of malignant transformation in PCN. The tumor diameter in the low-risk group was (34.72±12.35) mm, while that in the high-risk group was (41.26±15.78) mm, with a statistically significant difference between the two groups. Multivariate regression analysis identi-fied elevated CA19-9 levels, presence of clinical symptoms, tumor location in the pancreatic head or neck, and smoking history as inde-pendent risk factors. Based on these factors, a prediction model was constructed with an area under the receiver operating characteristiccurve (ROC curve) (AUC) of 0.83 [95%CI: (0.75, 0.90)]. The optimal cut-off value was calculated as 0.51, at which the model yielded asensitivity of 64.2%, a specificity of 87.1%, and an overall accuracy of 77.2%.Conclusions Elevated CA19-9 levels, presence of clini-cal symptoms, tumor location in the pancreatic head or neck, and smoking history are indicative of an increased risk of malignant trans-formation in PCN. Based on these key factors, developed a highly accurate prediction model, which has the potential to become an im-portant tool for assessing the nature of PCN and guiding personalized treatment in future clinical practice. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
