年轻女性ⅠA1~ⅡB期宫颈癌盆腔淋巴结转移相关因素的分析和预测

Analysis and Prediction of Related Factors of Pelvic Lymph Node Metastasis in Young Women with StageⅠA1-ⅡB Cervical Cancer

目的研究影响ⅠA1~ⅡB期年轻(≤40岁)宫颈癌患者盆腔淋巴结转移相关因素,建立风险预测模型,为临床实践提供参考。方法回顾性分析2012年1月—2019年10月收治于福建省妇幼保健院的169例ⅠA1~ⅡB期年轻宫颈癌病例的临床病理资料,用Logistic回归法筛选影响盆腔淋巴结转移相关因素,并建立风险预测模型和算出ROC曲线下面积。结果经过计数资料的单因素分析,结果显示,Ⅱ期、肿瘤长径>4 cm、脉管癌栓、肌层浸润、宫旁浸润组盆腔淋巴结转移率更高,而血红蛋白浓度>100 g/L组盆腔淋巴结转移率更低,差异有统计学意义(P<0.05);经过Logistic回归分析,纳入预报模型的因素为肿瘤长径、脉管癌栓、肌层浸润、宫旁浸润、血红蛋白浓度,该模型对训练样本预测的一致百分率为81.1%,该模型ROC曲线下面积为0.825,表明该预报模型预报能力良好。结论对于ⅠA1~ⅡB期年轻宫颈癌患者,肿瘤长径、脉管癌栓、肌层浸润、宫旁浸润、血红蛋白浓度是用于构建预测其盆腔淋巴结转移风险模型的相对最优因素,由此构建的模型具有良好的拟合优度和实用性。

Objective To study the related factors that affect pelvic lymph node metastasis in young(≤40 years old cervical cancer patients at stage ⅠA1-ⅡB, establish a risk prediction model, and provide reference for clinical practice.Methods Retrospectively analyzed the clinicopathological data of 169 young cervical cancer cases of stage ⅠA1-ⅡB who were admitted to Fujian Provincial Maternity and Child Health Hospital from January 2012 to October 2019. Logistic regression method was used to screen the related factors affecting pelvic lymph node metastasis, and establish risk prediction model and calculate the area under the ROC curve. Results After a single factor analysis of the enumeration data, the results showed that the pelvic lymph node metastasis rate was higher in the stage Ⅱ, tumor length>4 cm, vascular tumor thrombus, myometrial infiltration, and parauterine infiltration group, while the hemoglobin concentration >100 g/L group of pelvic lymph node metastasis rate was lower, and the difference was statistically significant(P<0.05);after Logistic regression analysis, the factors included in the prediction model were tumor length, vascular tumor thrombus, myometrial infiltration, parauterine infiltration, hemoglobin concentration, the model predicts a consistent percentage of 81.1% for the training samples, and the area under the ROC curve of the model was 0.825, indicating that the forecasting model has a good forecasting ability. Conclusion For young patients with stage ⅠA1-ⅡB cervical cancer, tumor length, vascular tumor thrombus, myometrial infiltration, parauterine infiltration, and hemoglobin concentration are the relatively optimal factors for constructing a model for predicting the risk of pelvic lymph node metastasis. The constructed model has fine goodness of fit and practicality.