淋巴细胞与单核细胞比值对早期宫颈癌患者盆腔淋巴结转移的预测价值

目的探讨淋巴细胞与单核细胞比值(LMR)对早期宫颈癌患者盆腔淋巴结转移(PLNM)的预测价值。 方法选择2008年1月至2018年1月，于川北医学院附属南充市中心医院初次接受宫颈癌根治性术的516例早期宫颈癌患者为研究对象。根据是否发生PLNM，将其分为PLNM(＋)组(n＝113)和PLNM(－)组(n＝403)。采用回顾性分析法，分析所有受试者的临床病例资料。采用χ²检验，对2组患者年龄、国际妇产科联盟(FIGO)临床分期，手术切除组织的病理学分级、组织病理学类型、肿瘤直径、血清鳞状细胞癌抗原(SCC-Ag)含量，术中是否发现脉管癌栓、宫旁浸润，宫颈间质浸润深度，以及LMR构成比等进行统计学比较。绘制LMR预测宫颈癌PLNM的受试者工作特征(ROC)曲线，对LMR预测早期宫颈癌患者发生PLNM进行分析，并计算ROC曲线下面积(ROC-AUC)。采用多因素非条件logistic回归分析，判断LMR预测早期宫颈癌患者发生PLNM的准确性。本研究遵循的程序符合2013年修订的《世界医学协会赫尔辛基宣言》要求。2组患者年龄比较，差异无统计学意义(P>0.05)。 结果①单因素分析结果：PLNM(＋)组与PLNM(－)组患者FIGO临床分期、手术切除组织的病理学分级、组织病理学类型、肿瘤直径、血清SCC-Ag含量、术中是否发现脉管癌栓及宫旁浸润、宫颈间质浸润深度、LMR构成比比较，差异均有统计学意义(χ²＝13.057、18.998、9.584、85.667、7.225、88.538、69.807、32.644、126.453，P<0.05)。②多因素非条件logistic回归分析结果：早期宫颈癌患者FIGO临床分期为Ⅱ_A期(OR＝2.805，95%CI：1.224～4.160，P＝0.009)，手术切除组织的病理学分级为G1(OR＝2.256，95%CI：1.245～6.323，P＝0.013)，脉管癌栓(OR＝5.818，95%CI：3.139～10.782，P<0.001)，宫旁浸润(OR＝4.482，95%CI：1.730～11.615，P<0.002)，肿瘤直径≥4 cm(OR＝3.955，95%CI：2.119～7.383，P<0.001)，以及血清SCC-Ag含量≥1.6 ng/mL(OR＝1.909，95%CI：1.057～3.446，P＝0.032)，均为早期宫颈癌患者发生PLNM的危险因素，而LMR>2.2(OR＝0.155，95%CI：0.081～0.296，P<0.001)，则为早期宫颈癌患者发生PLNM的保护因素。③ROC曲线分析结果：LMR预测早期宫颈癌患者PLNM的ROC-AUC为0.790(95%CI：0.753～0.825，P<0.001)。根据约登指数(Youden index)最大原则，LMR预测早期宫颈癌患者发生PLNM的最佳临界值为2.2，此时其预测早期宫颈癌患者发生PLNM的敏感度为66.4%，特异度为85.9%。 结论术前外周血LMR>2.2，可作为预测早期宫颈癌发生PLNM的保护因素。

Predictive value of preoperative lymphocyte-to-monocyte ratio on pelvic lymph node metastasis of early cervical cancer patients

Objective To explore the value of lymphocyte-to-monocyte ratio(LMR)on pelvic lymph node metastasis(PLNM)of early cervical cancer patients.Methods The clinical and pathological data of 516 patients with early cervical cancer whose initial treatment were radical hysterectomy in the Affiliated Nanchong Central Hospital of North Sichuan Medical College from January 2008 to January 2018 were selected into this study.According to being accompanied with PLNM or not,they were divided into PLNM(+)group(n=113)and PLNM(-)group(n=403).Retrospective analysis was used to analyze clinical data of all subjects.Chi-square was used to statistically compare the proportions of the age,International Federation of Gynecology and Obstetrics(FIGO)clinical stages,histopathological grades,histopathological types,vessel carcinoma embolus,parametrial invasion,cervical interstitial invasion depth,tumor diameters,levels of serum squamous cell carcinoma antigen(SCC-Ag),LMR between two groups.The receiver operating characteristic(ROC)curve of LMR predicting PLNM in early cervical cancer patients was drawn and analyzed,and the area under ROC curve(ROC-AUC)was calculated.The accuracy of LMR in the prediction of PLNM in early cervical cancer patients was evaluated by multivariate unconditional logistic regression analysis.This study met the requirements of the World Medical Association Declaration of Helsinki revised in 2013.There was no significant difference between two groups in age(P>0.05).Results①The results of univariate analysis showed that there were significant differences in proportions of FIGO clinical stages,histopathological grades,histopathological types,tumor diameters,levels of serum SCC-Ag,vessel carcinoma embolus,parametrial invasion,depth of cervical interstitial invasion and LMR(χ2=13.057,18.998,9.584,85.667,7.225,88.538,69.807,32.644,126.453;P<0.05).②The results of multivariate unconditional logistic regression analysis showed that FIGO clinical stageⅡA(OR=2.805,95%CI:1.224-4.160,P=0.009),histopathological grade G1(OR=2.256,95%CI:1.245-6.323,P=0.013),vessel carcinoma embolus(OR=5.818,95%CI:3.139-10.782,P<0.001),parametrial invasion(OR=4.482,95%CI:1.730-11.615,P<0.002),tumor diameters≥4 cm(OR=3.955,95%CI:2.119-7.383,P<0.001),levels of serum SCC-Ag≥1.6 ng/mL(OR=1.909,95%CI:1.057-3.446,P=0.032)were risk factors of PLNM in early cervical cancer patients,while LMR>2.2(OR=0.155,95%CI:0.081-0.296,P<0.001)was the protective factor of PLNM in early cervical cancer patients.③The ROC curve of LMR predicting PLNM of early cervical cancer showed that the ROC-AUC was 0.790(95%CI:0.753-0.825,P<0.001).According to the maximum principle of the Youden index,the optimal cut-off value of LMR predicting PLNM in patients with early cervical cancer was 2.2,and the sensitivity and specificity were 66.4%and 85.9%,respectively.Conclusion Peripheral blood LMR>2.2 is an independent protective factor to predict PLNM of early cervical cancer.