【摘要】 目的：通过建立术前膀胱癌淋巴结转移(LNM)列线图预测模型以支持临床决策，选择出最适合患者的淋巴结清扫术。方法：利用SEER数据库搜索并筛选出2010～2015年经病理诊断为膀胱癌的患者共7 109例，并按7∶3比例随机分为训练组(n=4 980)与验证组(n=2 129),采用组间差异分析和多因素Logistic回归分析筛选出膀胱癌淋巴结转移的独立危险因素并制作列线图模型。运用受试者工作特征(ROC)曲线、一致性指数、校准曲线验证列线图模型的区分度和一致性，决策曲线分析评估列线图模型的预测效能，并使用Bootstrap重采样1 000次对模型进行内部验证。结果：多因素Logistic分析显示，膀胱癌组织类型、病理分级、膀胱癌T、M分期、年龄及肿瘤大小是预测膀胱癌淋巴结转移的独立危险因素。基于以上6个指标构建列线图预测模型，其训练组和验证组一致性指数分别为0.869(95%CI:0.856～0.882)、0.865(95%CI:0.852～0.878)。校准曲线、临床决策曲线分别提示模型一致性和临床应用效能良好。结论：本研究构建的膀胱癌淋巴结转移列线图预测模型具有较好的预测精度，有助于临床医生为膀胱癌患者做出合适的临床决策。更多还原

【Abstract】 Objective To select the most suitable lymph node dissection for patients by establishing a lymph node metastasis(LNM) nomogram prediction model for bladder cancer to support clinical decision making.Method A total of 7 109 patients diagnosed with bladder cancer by pathology from 2010 to 2015 were searched and screened using the SEER database, and they were randomly divided into a training group(n=4 980) and a validation group(n=2 129) in a 7∶3 ratio, and independent risk factors for lymph node metastasis in bladder cancer were screened using between-group difference analysis and multi-factor logistic regression analysis and Nomogram model was produced. The ROC curve, consistency index, and calibration curve were used to verify the discrimination and consistency of the nomogram model, and the decision curve analysis was used to assess the predictive efficacy of the nomogram model, and the model was internally validated using Bootstrap resampling 1 000 times.Results Multi-factor logistic analysis showed that bladder cancer tissue type, pathological grade, T and M stages of bladder cancer, age and tumor size were independent risk factors for predicting lymph node metastasis in bladder cancer. The nomogram prediction model was constructed based on the above six indicators, and its consistency index was 0.869(95% CI:0.856～0.882) and 0.865(95% CI:0.852～0.878) in the training and validation groups, respectively. The calibration curve and clinical decision curve suggested good model consistency and clinical application efficacy, respectively.Conclusion The lymph node metastasis nomogram prediction model for bladder cancer constructed in this study has good prediction accuracy and helps clinicians to make appropriate clinical decisions for bladder cancer patients.更多还原