术前外周血多指标联合检测分析在乳腺癌诊断中应用价值

目的近年来乳腺癌的发病率逐年增加,早期诊断显得尤为重要。本研究分析术前外周血中性粒细胞和淋巴细胞比值(neutrophil to lymphocyte ratio,NLR)、血小板和淋巴细胞比值(platelet-to-lymphocyte ratio,PLR)、系统性免疫性炎症指数(systemic immune-inflammation index,SII)、衍生中性粒细胞与淋巴细胞比值(derived neutrophil-lymphocyte ratio,dNLR)和淋巴细胞/单核细胞比值(lymphocyte monocyte ratio,LMR)在乳腺良性疾病及乳腺癌中的差异,分析其诊断价值。方法回顾性分析安徽省第二人民医院2016-01-01-2019-01-01手术治疗的193例乳腺肿块患者临床资料,根据术后常规病理学检查结果分为乳腺良性疾病组和乳腺癌组,对每组病例所对应的NLR、PLR、SII、dNLR和LMR进行对比分析,并根据结果进行二元Logistic回归分析建立诊断模型,应用受试者工作特征(receiver operating characteristic,ROC)曲线鉴别乳腺良恶性肿块的最佳临界值,并分析与相关病理参数的关联。结果乳腺癌组NLR(t=5.411,P<0.001)、PLR(t=2.457,P=0.016)、SII(t=4.299,P<0.001)及dNLR(t=3.977,P<0.001)高于乳腺良性疾病组,差异有统计学意义;2组患者LMR对比差异无统计学意义,t=1.947,P=0.053。由ROC曲线分析得出,鉴别乳腺肿块良恶性的NLR最佳临界值为1.659,灵敏度为62.9%,特异度为77.2%,曲线下面积(area under curve,AUC)为0.742,在4组指标中诊断效能最高。利用二元Logistic回归分析法建立NLR+PLR+dNLR诊断模型,其最佳临界值为7.810,此时灵敏度为48.9%,特异度为93.5%,AUC为0.751;NLR及NLR+PLR+dNLR诊断模型与乳腺癌TNM分期相关,χ^2=8.820,P=0.012。结论术前外周血NLR及NLR+PLR+dNLR诊断模型对乳腺癌的诊断及病情评估具有指导意义。

Application value of preoperative peripheral blood multi-index combined detection and analysis in the diagnosis of breast cancer

OBJECTIVE In recent years,the incidence of breast cancer has been increasing year by year,and early diagnosis is particularly important.This study was to investigate the diagnostic value of preoperative NLR,PLR,SII,dNLR,LMR in the diagnosis of benign breast disease and breast cancer.METHODS A retrospective study was performed to analyze the clinical data of 193 patients with breast mass who accepted surgery treatment in the Anhui No.2 Provincial People’s Hospital from January 2016 to January 2019.The patients were divided into a benign breast disease group and breast cancer group according to postoperative pathological results.The differences of NLR,PLR,SII,dNLR and LMR between the two groups were discussed.Binary logistic regression analysis was performed to establish the diagnostic model.Furthermore,the optimal cut-off value to discriminate between benign breast disease and breast cancer was obtained via the receiver operator characteristic curve(ROC);the correlation with pathological parameters was also analyzed.RESULTS NLR,PLR,SII and dNLR were significantly different between the two groups(t=5.411,P<0.001;t=2.457,P=0.016;t=4.299,P<0.001;t=3.977,P<0.001).PLR was no significantly difference(t=1.947,P=0.053).The optimal NLR cut-off point obtained from ROC analysis was 1.659(sensitivity 62.9%,specificity 77.2%),while area under curve(AUC)was 0.742,among the four groups of indicators,the diagnostic efficiency was the highest.Establish the diagnosis model of NLR+PLR+dNLR by binary logistic regression analysis,the optimal NLR+PLR+dNLR cut-off point obtained from ROC analysis was 7.810(sensitivity 48.9%,specificity 93.5%),while AUC was 0.751.NLR and NLR+PLR+dNLR diagnostic model were significantly related with TNM stage of breast cancer(χ^2=8.820,P=0.012).CONCLUSION Preoperative peripheral blood NLR and NLR+PLR+dNLR diagnostic model can provide guidance for the diagnosis and evaluation of breast cancer.