摘要
摘要目的探讨阴茎鳞状细胞癌(SCCP)患者腹股沟淋巴结转移(LLM)的独立预测因素,并建立列线图模型对其LLM风险进行个体化预测。方法回顾性分析2009年7月至2019年6月空军军医大学西京医院确诊的101例SCCP患者的病例资料。年龄55(26~84)岁。25例(24.8%)可触及腹股沟淋巴结肿大,76例(75.2%)腹股沟淋巴结无肿大。肿瘤分期T1期47例(46.5%),T2期40例(39.6%),T3期14例(13.9%);Broder分级1级67例(66.3%),2级21例(20.8%),3级13例(12.9%)。实验室检查:纤维蛋白原含量2.84(1.72~5.00)g/L,碱性磷酸酶80(32~214)U/L,血红蛋白147(81~180)g/L,血小板计数(193.74±65.68)×109/L,中性粒细胞绝对值3.98(1.19~11.85)×109/L,单核细胞绝对值0.44(0.17~1.90)×109/L,淋巴细胞绝对值1.68(0.58~4.13)×109/L;血小板计数/淋巴细胞绝对值113.38(18.80~418.42),中性粒细胞绝对值/淋巴细胞绝对值2.42(0.59~10.22),淋巴细胞绝对值/单核细胞绝对值3.84(1.08~9.89)。术后病理结果提示腹股沟淋巴结转移26例(25.7%),无淋巴结转移75例(74.3%)。采用单因素和多因素logistic回归分析确定LLM的独立预测因素,根据多因素回归模型运用R软件绘制列线图。采用受试者工作特征(ROC)曲线下面积(AUC)评价列线图模型的区分度,并采用bootstrap法绘制校准图对模型的预测性能进行验证。结果单因素和多因素分析结果显示,临床腹股沟淋巴结状态(P=0.006)、T分期(P=0.021)、Broder分级(P=0.017)和中性粒细胞绝对值(P=0.043)是LLM的独立预测因素。列线图模型预测LLM风险的区分度为0.875(AUC=0.875,95%CI 0.859~0.891),且模型预测的LLM风险与实际LLM发生率具有良好的一致性,模型预测LLM风险的误差均在10%以内。结论腹股沟淋巴结状态、T分期、Broder分级和中性粒细胞绝对值是SCCP患者LLM的独立预测因素,基于上述指标建立的LLM风险预测模型有助于筛选需要行预防性腹股沟淋巴结清扫的SCCP患者。
Objective To investigate the independent predictors of inguinal lymph node metastasis(LLM)in patients with penile squamous cell carcinoma(SCCP),and to establish a nomogram for predicting individual LLM risk.Methods The data of patients with SCCP diagnosed at the department of urology,Xijing Hospital from July 2009 to June 2019 were analyzed retrospectively.A total of 101 patients were included in this study,with age of 55(26-84)years.There were 25(24.8%)and 76(75.2%)patients with and without palpable inguinal lymph node enlargement,respectively.There were 47 cases(46.5%),40 cases(39.6%)and 14 cases(13.9%)in T1,T2 and T3 stages,respectively;there were 67 cases(66.3%),21 cases(20.8%)and 13 cases(12.9%)in Broder 1,2 and 3,respectively.The average value(or median)of fibrinogen was 2.84(1.72-5.00)g/L;alkaline phosphatase was 80(32±214)U,hemoglobin was 147(81-180)g,platelet count was(193.74±65.68×109/L,absolute value of neutrophils,monocytes and lymphocytes were 3.98(1.19-11.85)×109/L,0.44(0.17-1.90)×109/L and 1.68(0.58-4.13)×109/L,respectively.The average(or median)value of PLR,NLR and LMR were 113.38(18.80-418.42),2.42(0.59-10.22)and 3.84(1.08-9.89),respectively.There were 26 cases(25.7%)with LLM and 75 cases(74.3%)without LLM.The independent predictors of LLM were identified by univariate and multivariate logistic regression analyses.The R software was used to establish the nomogram by integrating all independent predictors,and the bootstrap method was used to internally validated our nomogram,where the value of AUC(area under the ROC curve)was calculated and the calibration plot was drawn.Results Clinical inguinal lymph node status(P<0.006),T stage(P<0.021),Broder grade(P<0.017)and absolute neutrophil value(P<0.043)were independent predictors of LLM.The accuracy of our nomogram was 0.875(AUC=0.875,95%CI 0.859-0.891);Moreover,the risk of LLM predicted by nomogram was in good consistency with the actual LLM rate,and the errors of the nomogram-predicted LLM risks were all within 10%.Conclusions Clinical inguinal lymph node status,T stage,broder grade and absolute value of neutrophils were identified as independent predictors of LLM in patients with SCCP on the basis of single center data.A generic nomogram predicting LLM risk for Chinese patients was developed,which would be helpful to screen SCCP patients who need prophylactic inguinal lymph node dissection.
作者
侯广东
郑昱
王福利
李西安
张更
张龙龙
郑万祥
袁建林
Hou Guangdong;Zheng Yu;Wang Fuli;Li Xi'an;Zhang Geng;Zhang Longlong;Zheng Wanxiang;Yuan Jianlin(Department of Urology,Xijing Hospital of the Air Force Medical University,X i’an 710032,China)
出处
《中华泌尿外科杂志》
CAS
CSCD
北大核心
2019年第12期909-914,共6页
Chinese Journal of Urology
关键词
阴茎肿瘤
鳞状细胞癌
淋巴结转移
列线图
Penile neoplasms
Squamous cell carcinoma
Inguinal lymph node metastasis
Nomogram
