摘要
目的分析超声引导下细针穿刺细胞学检查(US-FNAC)低估乳腺导管原位癌(DCIS)的影响因素,并以此构建列线图预测模型并进行验证。方法回顾性选取宜兴市中医医院2020年1月至2023年6月收治的105例经US-FNAC诊断为DCIS的患者,均为女性,年龄40~76岁。根据患者术后病理诊断结果,将其分为被低估组(27例)和未被低估组(78例)。收集患者临床相关资料,分析影响DCIS被低估的影响因素,并以此构建列线图预测模型,绘制受试者操作特征(ROC)曲线,用曲线下面积(AUC)分析预测模型对DCIS术前被低估风险的预测效能。结果单因素分析显示,被低估组可触及肿块占比、肿瘤大小>2 cm占比、伴微钙化占比、乳腺影像报告与数据系统(BI-RADS)分类为4B~5级占比均高于未被低估组(P<0.05)。二元Logistic回归分析显示,肿瘤大小(OR=4.453,95%CI:1.890~10.486)、伴微钙化(OR=3.079,95%CI:1.650~5.742)、BI-RADS分类(OR=5.211,95%CI:2.528~10.740)是影响DCIS术前被低估的独立危险因素(P<0.05)。基于上述影响因素构建的列线图预测模型经Bootstrap法内部验证显示,C-index为0.865(95%CI:0.785~0.952),预测DCIS术前被低估的校正曲线趋近于理想曲线(P>0.05)。ROC曲线结果显示,构建的列线图模型预测DCIS术前被低估的灵敏度为87.90%、特异度为88.60%,AUC为0.895(95%CI:0.823~0.966)。结论肿瘤大小、伴微钙化、BI-RADS分类是影响DCIS术前被低估的独立危险因素,基于此构建的列线图预测模型可较好地评估DCIS术前被低估的发生风险。
Objective To analyze the underestimated influencing factors of ductal carcinoma in situ(DCIS)detected by ultrasound-guided fine needle aspiration cytology(US-FNAC),and to construct and validate the nomogram prediction model.Methods A total of 105 patients with DCIS diagnosed by US-FNAC in Yixing Hospital of Traditional Chinese Medicine from January 2020 to June 2023 were retrospectively selected.All patients were female and aged 40-76 years.According to postoperative pathological results,they were assigned to underestimated group(n=27)and non-underestimated group(n=78).The clinical data of the patients were collected to analyze the influencing factors of the underestimation of DCIS,and a nomogram prediction model was constructed.The receiver operating characteristic curve(ROC)was drawn,and the area under the curve(AUC)was used to analyze the predictive efficacy of the prediction model for the underestimated risk of DCIS before operation.Results Univariate analysis showed that the proportions of palpable masses,tumor size>2 cm,microcalcification,and breast imaging reporting and data system(BI-RADS)4B-5 in the underestimated group were higher than those in the non-underestimated group(P<0.05).Binary logistic regression analysis showed that tumor size[(odds ratio,OR)=4.453,95%(confidence interval,CI):1.890-10.486],microcalcification(OR=3.079,95%CI:1.650-5.742),BI-RADS classification(OR=5.211,95%CI:2.528-10.740)were the independent risk factors for preoperative underestimation of DCIS(P<0.05).Internal validation of the nomogram prediction model based on the above factors showed that the C-index was 0.865(95%CI:0.785-0.952),and the correction curve for predicting the underestimation of DCIS was close to the ideal curve(P>0.05).ROC curve showed that the nomogram model had a sensitivity of 87.90%,a specificity of 88.60%,and an AUC of 0.895(95%CI:0.823-0.966).Conclusion Tumor size,microcalcification and BI-RADS classification are independent risk factors for preoperative underestimation of DCIS.The nomogram prediction model based on these factors can better evaluate the risk of preoperative underestimation of DCIS.
作者
王人杰
陈立丰
潘志平
沈莉敏
Wang Renjie;Chen Lifeng;Pan Zhiping;Shen Limin(Department of Ultrasound,Yixing Hospital of Traditional Chinese Medicine,Wuxi 214200,China)
出处
《海军医学杂志》
2025年第8期793-797,共5页
Journal of Navy Medicine
基金
江苏省科技基础研究计划自然科学基金项目(BK20221202)。
